rocksdb/db/version_set.cc

2802 lines
97 KiB
C++
Raw Normal View History

// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
//
// Copyright (c) 2011 The LevelDB Authors. All rights reserved.
// Use of this source code is governed by a BSD-style license that can be
// found in the LICENSE file. See the AUTHORS file for names of contributors.
#include "db/version_set.h"
#ifndef __STDC_FORMAT_MACROS
#define __STDC_FORMAT_MACROS
#endif
#include <inttypes.h>
#include <stdio.h>
#include <algorithm>
#include <map>
2014-01-30 00:26:43 +01:00
#include <set>
#include <climits>
#include <unordered_map>
#include <vector>
#include <string>
#include "db/filename.h"
#include "db/log_reader.h"
#include "db/log_writer.h"
#include "db/memtable.h"
#include "db/merge_context.h"
#include "db/table_cache.h"
#include "db/compaction.h"
#include "db/version_builder.h"
#include "db/writebuffer.h"
#include "rocksdb/env.h"
#include "rocksdb/merge_operator.h"
#include "table/table_reader.h"
#include "table/merger.h"
#include "table/two_level_iterator.h"
#include "table/format.h"
#include "table/plain_table_factory.h"
#include "table/meta_blocks.h"
#include "table/get_context.h"
#include "util/coding.h"
#include "util/logging.h"
#include "util/stop_watch.h"
namespace rocksdb {
namespace {
// Find File in LevelFilesBrief data structure
// Within an index range defined by left and right
int FindFileInRange(const InternalKeyComparator& icmp,
const LevelFilesBrief& file_level,
const Slice& key,
uint32_t left,
uint32_t right) {
while (left < right) {
uint32_t mid = (left + right) / 2;
const FdWithKeyRange& f = file_level.files[mid];
if (icmp.InternalKeyComparator::Compare(f.largest_key, key) < 0) {
// Key at "mid.largest" is < "target". Therefore all
// files at or before "mid" are uninteresting.
left = mid + 1;
} else {
// Key at "mid.largest" is >= "target". Therefore all files
// after "mid" are uninteresting.
right = mid;
}
}
return right;
}
// Class to help choose the next file to search for the particular key.
// Searches and returns files level by level.
// We can search level-by-level since entries never hop across
// levels. Therefore we are guaranteed that if we find data
// in a smaller level, later levels are irrelevant (unless we
// are MergeInProgress).
class FilePicker {
public:
FilePicker(
std::vector<FileMetaData*>* files,
const Slice& user_key,
const Slice& ikey,
autovector<LevelFilesBrief>* file_levels,
unsigned int num_levels,
FileIndexer* file_indexer,
const Comparator* user_comparator,
const InternalKeyComparator* internal_comparator)
: num_levels_(num_levels),
curr_level_(-1),
search_left_bound_(0),
search_right_bound_(FileIndexer::kLevelMaxIndex),
#ifndef NDEBUG
files_(files),
#endif
level_files_brief_(file_levels),
user_key_(user_key),
ikey_(ikey),
file_indexer_(file_indexer),
user_comparator_(user_comparator),
internal_comparator_(internal_comparator) {
// Setup member variables to search first level.
search_ended_ = !PrepareNextLevel();
if (!search_ended_) {
// Prefetch Level 0 table data to avoid cache miss if possible.
for (unsigned int i = 0; i < (*level_files_brief_)[0].num_files; ++i) {
auto* r = (*level_files_brief_)[0].files[i].fd.table_reader;
if (r) {
r->Prepare(ikey);
}
}
}
}
FdWithKeyRange* GetNextFile() {
while (!search_ended_) { // Loops over different levels.
while (curr_index_in_curr_level_ < curr_file_level_->num_files) {
// Loops over all files in current level.
FdWithKeyRange* f = &curr_file_level_->files[curr_index_in_curr_level_];
int cmp_largest = -1;
// Do key range filtering of files or/and fractional cascading if:
// (1) not all the files are in level 0, or
// (2) there are more than 3 Level 0 files
// If there are only 3 or less level 0 files in the system, we skip
// the key range filtering. In this case, more likely, the system is
// highly tuned to minimize number of tables queried by each query,
// so it is unlikely that key range filtering is more efficient than
// querying the files.
if (num_levels_ > 1 || curr_file_level_->num_files > 3) {
// Check if key is within a file's range. If search left bound and
// right bound point to the same find, we are sure key falls in
// range.
assert(
curr_level_ == 0 ||
curr_index_in_curr_level_ == start_index_in_curr_level_ ||
user_comparator_->Compare(user_key_,
ExtractUserKey(f->smallest_key)) <= 0);
int cmp_smallest = user_comparator_->Compare(user_key_,
ExtractUserKey(f->smallest_key));
if (cmp_smallest >= 0) {
cmp_largest = user_comparator_->Compare(user_key_,
ExtractUserKey(f->largest_key));
}
// Setup file search bound for the next level based on the
// comparison results
if (curr_level_ > 0) {
file_indexer_->GetNextLevelIndex(curr_level_,
curr_index_in_curr_level_,
cmp_smallest, cmp_largest,
&search_left_bound_,
&search_right_bound_);
}
// Key falls out of current file's range
if (cmp_smallest < 0 || cmp_largest > 0) {
if (curr_level_ == 0) {
++curr_index_in_curr_level_;
continue;
} else {
// Search next level.
break;
}
}
}
#ifndef NDEBUG
// Sanity check to make sure that the files are correctly sorted
if (prev_file_) {
if (curr_level_ != 0) {
int comp_sign = internal_comparator_->Compare(
prev_file_->largest_key, f->smallest_key);
assert(comp_sign < 0);
} else {
// level == 0, the current file cannot be newer than the previous
// one. Use compressed data structure, has no attribute seqNo
assert(curr_index_in_curr_level_ > 0);
assert(!NewestFirstBySeqNo(files_[0][curr_index_in_curr_level_],
files_[0][curr_index_in_curr_level_-1]));
}
}
prev_file_ = f;
#endif
if (curr_level_ > 0 && cmp_largest < 0) {
// No more files to search in this level.
search_ended_ = !PrepareNextLevel();
} else {
++curr_index_in_curr_level_;
}
return f;
}
// Start searching next level.
search_ended_ = !PrepareNextLevel();
}
// Search ended.
return nullptr;
}
private:
unsigned int num_levels_;
unsigned int curr_level_;
int32_t search_left_bound_;
int32_t search_right_bound_;
#ifndef NDEBUG
std::vector<FileMetaData*>* files_;
#endif
autovector<LevelFilesBrief>* level_files_brief_;
bool search_ended_;
LevelFilesBrief* curr_file_level_;
unsigned int curr_index_in_curr_level_;
unsigned int start_index_in_curr_level_;
Slice user_key_;
Slice ikey_;
FileIndexer* file_indexer_;
const Comparator* user_comparator_;
const InternalKeyComparator* internal_comparator_;
#ifndef NDEBUG
FdWithKeyRange* prev_file_;
#endif
// Setup local variables to search next level.
// Returns false if there are no more levels to search.
bool PrepareNextLevel() {
curr_level_++;
while (curr_level_ < num_levels_) {
curr_file_level_ = &(*level_files_brief_)[curr_level_];
if (curr_file_level_->num_files == 0) {
// When current level is empty, the search bound generated from upper
// level must be [0, -1] or [0, FileIndexer::kLevelMaxIndex] if it is
// also empty.
assert(search_left_bound_ == 0);
assert(search_right_bound_ == -1 ||
search_right_bound_ == FileIndexer::kLevelMaxIndex);
// Since current level is empty, it will need to search all files in
// the next level
search_left_bound_ = 0;
search_right_bound_ = FileIndexer::kLevelMaxIndex;
curr_level_++;
continue;
}
// Some files may overlap each other. We find
// all files that overlap user_key and process them in order from
// newest to oldest. In the context of merge-operator, this can occur at
// any level. Otherwise, it only occurs at Level-0 (since Put/Deletes
// are always compacted into a single entry).
int32_t start_index;
if (curr_level_ == 0) {
// On Level-0, we read through all files to check for overlap.
start_index = 0;
} else {
// On Level-n (n>=1), files are sorted. Binary search to find the
// earliest file whose largest key >= ikey. Search left bound and
// right bound are used to narrow the range.
if (search_left_bound_ == search_right_bound_) {
start_index = search_left_bound_;
} else if (search_left_bound_ < search_right_bound_) {
if (search_right_bound_ == FileIndexer::kLevelMaxIndex) {
search_right_bound_ =
static_cast<int32_t>(curr_file_level_->num_files) - 1;
}
start_index =
FindFileInRange(*internal_comparator_, *curr_file_level_, ikey_,
static_cast<uint32_t>(search_left_bound_),
static_cast<uint32_t>(search_right_bound_));
} else {
// search_left_bound > search_right_bound, key does not exist in
// this level. Since no comparision is done in this level, it will
// need to search all files in the next level.
search_left_bound_ = 0;
search_right_bound_ = FileIndexer::kLevelMaxIndex;
curr_level_++;
continue;
}
}
start_index_in_curr_level_ = start_index;
curr_index_in_curr_level_ = start_index;
#ifndef NDEBUG
prev_file_ = nullptr;
#endif
return true;
}
// curr_level_ = num_levels_. So, no more levels to search.
return false;
}
};
} // anonymous namespace
VersionStorageInfo::~VersionStorageInfo() { delete[] files_; }
Version::~Version() {
assert(refs_ == 0);
// Remove from linked list
prev_->next_ = next_;
next_->prev_ = prev_;
// Drop references to files
for (int level = 0; level < storage_info_.num_levels_; level++) {
for (size_t i = 0; i < storage_info_.files_[level].size(); i++) {
FileMetaData* f = storage_info_.files_[level][i];
assert(f->refs > 0);
f->refs--;
if (f->refs <= 0) {
if (f->table_reader_handle) {
cfd_->table_cache()->ReleaseHandle(f->table_reader_handle);
f->table_reader_handle = nullptr;
}
Speed up FindObsoleteFiles() Summary: There are two versions of FindObsoleteFiles(): * full scan, which is executed every 6 hours (and it's terribly slow) * no full scan, which is executed every time a background process finishes and iterator is deleted This diff is optimizing the second case (no full scan). Here's what we do before the diff: * Get the list of obsolete files (files with ref==0). Some files in obsolete_files set might actually be live. * Get the list of live files to avoid deleting files that are live. * Delete files that are in obsolete_files and not in live_files. After this diff: * The only files with ref==0 that are still live are files that have been part of move compaction. Don't include moved files in obsolete_files. * Get the list of obsolete files (which exclude moved files). * No need to get the list of live files, since all files in obsolete_files need to be deleted. I'll post the benchmark results, but you can get the feel of it here: https://reviews.facebook.net/D30123 This depends on D30123. P.S. We should do full scan only in failure scenarios, not every 6 hours. I'll do this in a follow-up diff. Test Plan: One new unit test. Made sure that unit test fails if we don't have a `if (!f->moved)` safeguard in ~Version. make check Big number of compactions and flushes: ./db_stress --threads=30 --ops_per_thread=20000000 --max_key=10000 --column_families=20 --clear_column_family_one_in=10000000 --verify_before_write=0 --reopen=15 --max_background_compactions=10 --max_background_flushes=10 --db=/fast-rocksdb-tmp/db_stress --prefixpercent=0 --iterpercent=0 --writepercent=75 --db_write_buffer_size=2000000 Reviewers: yhchiang, rven, sdong Reviewed By: sdong Subscribers: dhruba, leveldb Differential Revision: https://reviews.facebook.net/D30249
2014-12-22 12:04:45 +01:00
if (!f->moved) {
vset_->obsolete_files_.push_back(f);
} else {
// moved!
// TODO(icanadi) delete this outside of mutex
delete f;
}
}
}
}
}
create compressed_levels_ in Version, allocate its space using arena. Make Version::Get, Version::FindFile faster Summary: Define CompressedFileMetaData that just contains fd, smallest_slice, largest_slice. Create compressed_levels_ in Version, the space is allocated using arena Thus increase the file meta data locality, speed up "Get" and "FindFile" benchmark with in-memory tmpfs, could have 4% improvement under "random read" and 2% improvement under "read while writing" benchmark command: ./db_bench --db=/mnt/db/rocksdb --num_levels=6 --key_size=20 --prefix_size=20 --keys_per_prefix=0 --value_size=100 --block_size=4096 --cache_size=17179869184 --cache_numshardbits=6 --compression_type=none --compression_ratio=1 --min_level_to_compress=-1 --disable_seek_compaction=1 --hard_rate_limit=2 --write_buffer_size=134217728 --max_write_buffer_number=2 --level0_file_num_compaction_trigger=8 --target_file_size_base=33554432 --max_bytes_for_level_base=1073741824 --disable_wal=0 --sync=0 --disable_data_sync=1 --verify_checksum=1 --delete_obsolete_files_period_micros=314572800 --max_grandparent_overlap_factor=10 --max_background_compactions=4 --max_background_flushes=0 --level0_slowdown_writes_trigger=16 --level0_stop_writes_trigger=24 --statistics=0 --stats_per_interval=0 --stats_interval=1048576 --histogram=0 --use_plain_table=1 --open_files=-1 --mmap_read=1 --mmap_write=0 --memtablerep=prefix_hash --bloom_bits=10 --bloom_locality=1 --perf_level=0 --benchmarks=readwhilewriting,readwhilewriting,readwhilewriting --use_existing_db=1 --num=52428800 --threads=1 —writes_per_second=81920 Read Random: From 1.8363 ms/op, improve to 1.7587 ms/op. Read while writing: From 2.985 ms/op, improve to 2.924 ms/op. Test Plan: make all check Reviewers: ljin, haobo, yhchiang, sdong Reviewed By: sdong Subscribers: dhruba, igor Differential Revision: https://reviews.facebook.net/D19419
2014-07-10 07:14:39 +02:00
int FindFile(const InternalKeyComparator& icmp,
const LevelFilesBrief& file_level,
create compressed_levels_ in Version, allocate its space using arena. Make Version::Get, Version::FindFile faster Summary: Define CompressedFileMetaData that just contains fd, smallest_slice, largest_slice. Create compressed_levels_ in Version, the space is allocated using arena Thus increase the file meta data locality, speed up "Get" and "FindFile" benchmark with in-memory tmpfs, could have 4% improvement under "random read" and 2% improvement under "read while writing" benchmark command: ./db_bench --db=/mnt/db/rocksdb --num_levels=6 --key_size=20 --prefix_size=20 --keys_per_prefix=0 --value_size=100 --block_size=4096 --cache_size=17179869184 --cache_numshardbits=6 --compression_type=none --compression_ratio=1 --min_level_to_compress=-1 --disable_seek_compaction=1 --hard_rate_limit=2 --write_buffer_size=134217728 --max_write_buffer_number=2 --level0_file_num_compaction_trigger=8 --target_file_size_base=33554432 --max_bytes_for_level_base=1073741824 --disable_wal=0 --sync=0 --disable_data_sync=1 --verify_checksum=1 --delete_obsolete_files_period_micros=314572800 --max_grandparent_overlap_factor=10 --max_background_compactions=4 --max_background_flushes=0 --level0_slowdown_writes_trigger=16 --level0_stop_writes_trigger=24 --statistics=0 --stats_per_interval=0 --stats_interval=1048576 --histogram=0 --use_plain_table=1 --open_files=-1 --mmap_read=1 --mmap_write=0 --memtablerep=prefix_hash --bloom_bits=10 --bloom_locality=1 --perf_level=0 --benchmarks=readwhilewriting,readwhilewriting,readwhilewriting --use_existing_db=1 --num=52428800 --threads=1 —writes_per_second=81920 Read Random: From 1.8363 ms/op, improve to 1.7587 ms/op. Read while writing: From 2.985 ms/op, improve to 2.924 ms/op. Test Plan: make all check Reviewers: ljin, haobo, yhchiang, sdong Reviewed By: sdong Subscribers: dhruba, igor Differential Revision: https://reviews.facebook.net/D19419
2014-07-10 07:14:39 +02:00
const Slice& key) {
return FindFileInRange(icmp, file_level, key, 0,
static_cast<uint32_t>(file_level.num_files));
create compressed_levels_ in Version, allocate its space using arena. Make Version::Get, Version::FindFile faster Summary: Define CompressedFileMetaData that just contains fd, smallest_slice, largest_slice. Create compressed_levels_ in Version, the space is allocated using arena Thus increase the file meta data locality, speed up "Get" and "FindFile" benchmark with in-memory tmpfs, could have 4% improvement under "random read" and 2% improvement under "read while writing" benchmark command: ./db_bench --db=/mnt/db/rocksdb --num_levels=6 --key_size=20 --prefix_size=20 --keys_per_prefix=0 --value_size=100 --block_size=4096 --cache_size=17179869184 --cache_numshardbits=6 --compression_type=none --compression_ratio=1 --min_level_to_compress=-1 --disable_seek_compaction=1 --hard_rate_limit=2 --write_buffer_size=134217728 --max_write_buffer_number=2 --level0_file_num_compaction_trigger=8 --target_file_size_base=33554432 --max_bytes_for_level_base=1073741824 --disable_wal=0 --sync=0 --disable_data_sync=1 --verify_checksum=1 --delete_obsolete_files_period_micros=314572800 --max_grandparent_overlap_factor=10 --max_background_compactions=4 --max_background_flushes=0 --level0_slowdown_writes_trigger=16 --level0_stop_writes_trigger=24 --statistics=0 --stats_per_interval=0 --stats_interval=1048576 --histogram=0 --use_plain_table=1 --open_files=-1 --mmap_read=1 --mmap_write=0 --memtablerep=prefix_hash --bloom_bits=10 --bloom_locality=1 --perf_level=0 --benchmarks=readwhilewriting,readwhilewriting,readwhilewriting --use_existing_db=1 --num=52428800 --threads=1 —writes_per_second=81920 Read Random: From 1.8363 ms/op, improve to 1.7587 ms/op. Read while writing: From 2.985 ms/op, improve to 2.924 ms/op. Test Plan: make all check Reviewers: ljin, haobo, yhchiang, sdong Reviewed By: sdong Subscribers: dhruba, igor Differential Revision: https://reviews.facebook.net/D19419
2014-07-10 07:14:39 +02:00
}
void DoGenerateLevelFilesBrief(LevelFilesBrief* file_level,
const std::vector<FileMetaData*>& files,
Arena* arena) {
assert(file_level);
assert(files.size() >= 0);
assert(arena);
size_t num = files.size();
file_level->num_files = num;
char* mem = arena->AllocateAligned(num * sizeof(FdWithKeyRange));
file_level->files = new (mem)FdWithKeyRange[num];
for (size_t i = 0; i < num; i++) {
Slice smallest_key = files[i]->smallest.Encode();
Slice largest_key = files[i]->largest.Encode();
// Copy key slice to sequential memory
size_t smallest_size = smallest_key.size();
size_t largest_size = largest_key.size();
mem = arena->AllocateAligned(smallest_size + largest_size);
memcpy(mem, smallest_key.data(), smallest_size);
memcpy(mem + smallest_size, largest_key.data(), largest_size);
FdWithKeyRange& f = file_level->files[i];
f.fd = files[i]->fd;
f.smallest_key = Slice(mem, smallest_size);
f.largest_key = Slice(mem + smallest_size, largest_size);
}
}
static bool AfterFile(const Comparator* ucmp,
create compressed_levels_ in Version, allocate its space using arena. Make Version::Get, Version::FindFile faster Summary: Define CompressedFileMetaData that just contains fd, smallest_slice, largest_slice. Create compressed_levels_ in Version, the space is allocated using arena Thus increase the file meta data locality, speed up "Get" and "FindFile" benchmark with in-memory tmpfs, could have 4% improvement under "random read" and 2% improvement under "read while writing" benchmark command: ./db_bench --db=/mnt/db/rocksdb --num_levels=6 --key_size=20 --prefix_size=20 --keys_per_prefix=0 --value_size=100 --block_size=4096 --cache_size=17179869184 --cache_numshardbits=6 --compression_type=none --compression_ratio=1 --min_level_to_compress=-1 --disable_seek_compaction=1 --hard_rate_limit=2 --write_buffer_size=134217728 --max_write_buffer_number=2 --level0_file_num_compaction_trigger=8 --target_file_size_base=33554432 --max_bytes_for_level_base=1073741824 --disable_wal=0 --sync=0 --disable_data_sync=1 --verify_checksum=1 --delete_obsolete_files_period_micros=314572800 --max_grandparent_overlap_factor=10 --max_background_compactions=4 --max_background_flushes=0 --level0_slowdown_writes_trigger=16 --level0_stop_writes_trigger=24 --statistics=0 --stats_per_interval=0 --stats_interval=1048576 --histogram=0 --use_plain_table=1 --open_files=-1 --mmap_read=1 --mmap_write=0 --memtablerep=prefix_hash --bloom_bits=10 --bloom_locality=1 --perf_level=0 --benchmarks=readwhilewriting,readwhilewriting,readwhilewriting --use_existing_db=1 --num=52428800 --threads=1 —writes_per_second=81920 Read Random: From 1.8363 ms/op, improve to 1.7587 ms/op. Read while writing: From 2.985 ms/op, improve to 2.924 ms/op. Test Plan: make all check Reviewers: ljin, haobo, yhchiang, sdong Reviewed By: sdong Subscribers: dhruba, igor Differential Revision: https://reviews.facebook.net/D19419
2014-07-10 07:14:39 +02:00
const Slice* user_key, const FdWithKeyRange* f) {
// nullptr user_key occurs before all keys and is therefore never after *f
return (user_key != nullptr &&
create compressed_levels_ in Version, allocate its space using arena. Make Version::Get, Version::FindFile faster Summary: Define CompressedFileMetaData that just contains fd, smallest_slice, largest_slice. Create compressed_levels_ in Version, the space is allocated using arena Thus increase the file meta data locality, speed up "Get" and "FindFile" benchmark with in-memory tmpfs, could have 4% improvement under "random read" and 2% improvement under "read while writing" benchmark command: ./db_bench --db=/mnt/db/rocksdb --num_levels=6 --key_size=20 --prefix_size=20 --keys_per_prefix=0 --value_size=100 --block_size=4096 --cache_size=17179869184 --cache_numshardbits=6 --compression_type=none --compression_ratio=1 --min_level_to_compress=-1 --disable_seek_compaction=1 --hard_rate_limit=2 --write_buffer_size=134217728 --max_write_buffer_number=2 --level0_file_num_compaction_trigger=8 --target_file_size_base=33554432 --max_bytes_for_level_base=1073741824 --disable_wal=0 --sync=0 --disable_data_sync=1 --verify_checksum=1 --delete_obsolete_files_period_micros=314572800 --max_grandparent_overlap_factor=10 --max_background_compactions=4 --max_background_flushes=0 --level0_slowdown_writes_trigger=16 --level0_stop_writes_trigger=24 --statistics=0 --stats_per_interval=0 --stats_interval=1048576 --histogram=0 --use_plain_table=1 --open_files=-1 --mmap_read=1 --mmap_write=0 --memtablerep=prefix_hash --bloom_bits=10 --bloom_locality=1 --perf_level=0 --benchmarks=readwhilewriting,readwhilewriting,readwhilewriting --use_existing_db=1 --num=52428800 --threads=1 —writes_per_second=81920 Read Random: From 1.8363 ms/op, improve to 1.7587 ms/op. Read while writing: From 2.985 ms/op, improve to 2.924 ms/op. Test Plan: make all check Reviewers: ljin, haobo, yhchiang, sdong Reviewed By: sdong Subscribers: dhruba, igor Differential Revision: https://reviews.facebook.net/D19419
2014-07-10 07:14:39 +02:00
ucmp->Compare(*user_key, ExtractUserKey(f->largest_key)) > 0);
}
static bool BeforeFile(const Comparator* ucmp,
create compressed_levels_ in Version, allocate its space using arena. Make Version::Get, Version::FindFile faster Summary: Define CompressedFileMetaData that just contains fd, smallest_slice, largest_slice. Create compressed_levels_ in Version, the space is allocated using arena Thus increase the file meta data locality, speed up "Get" and "FindFile" benchmark with in-memory tmpfs, could have 4% improvement under "random read" and 2% improvement under "read while writing" benchmark command: ./db_bench --db=/mnt/db/rocksdb --num_levels=6 --key_size=20 --prefix_size=20 --keys_per_prefix=0 --value_size=100 --block_size=4096 --cache_size=17179869184 --cache_numshardbits=6 --compression_type=none --compression_ratio=1 --min_level_to_compress=-1 --disable_seek_compaction=1 --hard_rate_limit=2 --write_buffer_size=134217728 --max_write_buffer_number=2 --level0_file_num_compaction_trigger=8 --target_file_size_base=33554432 --max_bytes_for_level_base=1073741824 --disable_wal=0 --sync=0 --disable_data_sync=1 --verify_checksum=1 --delete_obsolete_files_period_micros=314572800 --max_grandparent_overlap_factor=10 --max_background_compactions=4 --max_background_flushes=0 --level0_slowdown_writes_trigger=16 --level0_stop_writes_trigger=24 --statistics=0 --stats_per_interval=0 --stats_interval=1048576 --histogram=0 --use_plain_table=1 --open_files=-1 --mmap_read=1 --mmap_write=0 --memtablerep=prefix_hash --bloom_bits=10 --bloom_locality=1 --perf_level=0 --benchmarks=readwhilewriting,readwhilewriting,readwhilewriting --use_existing_db=1 --num=52428800 --threads=1 —writes_per_second=81920 Read Random: From 1.8363 ms/op, improve to 1.7587 ms/op. Read while writing: From 2.985 ms/op, improve to 2.924 ms/op. Test Plan: make all check Reviewers: ljin, haobo, yhchiang, sdong Reviewed By: sdong Subscribers: dhruba, igor Differential Revision: https://reviews.facebook.net/D19419
2014-07-10 07:14:39 +02:00
const Slice* user_key, const FdWithKeyRange* f) {
// nullptr user_key occurs after all keys and is therefore never before *f
return (user_key != nullptr &&
create compressed_levels_ in Version, allocate its space using arena. Make Version::Get, Version::FindFile faster Summary: Define CompressedFileMetaData that just contains fd, smallest_slice, largest_slice. Create compressed_levels_ in Version, the space is allocated using arena Thus increase the file meta data locality, speed up "Get" and "FindFile" benchmark with in-memory tmpfs, could have 4% improvement under "random read" and 2% improvement under "read while writing" benchmark command: ./db_bench --db=/mnt/db/rocksdb --num_levels=6 --key_size=20 --prefix_size=20 --keys_per_prefix=0 --value_size=100 --block_size=4096 --cache_size=17179869184 --cache_numshardbits=6 --compression_type=none --compression_ratio=1 --min_level_to_compress=-1 --disable_seek_compaction=1 --hard_rate_limit=2 --write_buffer_size=134217728 --max_write_buffer_number=2 --level0_file_num_compaction_trigger=8 --target_file_size_base=33554432 --max_bytes_for_level_base=1073741824 --disable_wal=0 --sync=0 --disable_data_sync=1 --verify_checksum=1 --delete_obsolete_files_period_micros=314572800 --max_grandparent_overlap_factor=10 --max_background_compactions=4 --max_background_flushes=0 --level0_slowdown_writes_trigger=16 --level0_stop_writes_trigger=24 --statistics=0 --stats_per_interval=0 --stats_interval=1048576 --histogram=0 --use_plain_table=1 --open_files=-1 --mmap_read=1 --mmap_write=0 --memtablerep=prefix_hash --bloom_bits=10 --bloom_locality=1 --perf_level=0 --benchmarks=readwhilewriting,readwhilewriting,readwhilewriting --use_existing_db=1 --num=52428800 --threads=1 —writes_per_second=81920 Read Random: From 1.8363 ms/op, improve to 1.7587 ms/op. Read while writing: From 2.985 ms/op, improve to 2.924 ms/op. Test Plan: make all check Reviewers: ljin, haobo, yhchiang, sdong Reviewed By: sdong Subscribers: dhruba, igor Differential Revision: https://reviews.facebook.net/D19419
2014-07-10 07:14:39 +02:00
ucmp->Compare(*user_key, ExtractUserKey(f->smallest_key)) < 0);
}
bool SomeFileOverlapsRange(
const InternalKeyComparator& icmp,
bool disjoint_sorted_files,
const LevelFilesBrief& file_level,
const Slice* smallest_user_key,
const Slice* largest_user_key) {
const Comparator* ucmp = icmp.user_comparator();
if (!disjoint_sorted_files) {
// Need to check against all files
create compressed_levels_ in Version, allocate its space using arena. Make Version::Get, Version::FindFile faster Summary: Define CompressedFileMetaData that just contains fd, smallest_slice, largest_slice. Create compressed_levels_ in Version, the space is allocated using arena Thus increase the file meta data locality, speed up "Get" and "FindFile" benchmark with in-memory tmpfs, could have 4% improvement under "random read" and 2% improvement under "read while writing" benchmark command: ./db_bench --db=/mnt/db/rocksdb --num_levels=6 --key_size=20 --prefix_size=20 --keys_per_prefix=0 --value_size=100 --block_size=4096 --cache_size=17179869184 --cache_numshardbits=6 --compression_type=none --compression_ratio=1 --min_level_to_compress=-1 --disable_seek_compaction=1 --hard_rate_limit=2 --write_buffer_size=134217728 --max_write_buffer_number=2 --level0_file_num_compaction_trigger=8 --target_file_size_base=33554432 --max_bytes_for_level_base=1073741824 --disable_wal=0 --sync=0 --disable_data_sync=1 --verify_checksum=1 --delete_obsolete_files_period_micros=314572800 --max_grandparent_overlap_factor=10 --max_background_compactions=4 --max_background_flushes=0 --level0_slowdown_writes_trigger=16 --level0_stop_writes_trigger=24 --statistics=0 --stats_per_interval=0 --stats_interval=1048576 --histogram=0 --use_plain_table=1 --open_files=-1 --mmap_read=1 --mmap_write=0 --memtablerep=prefix_hash --bloom_bits=10 --bloom_locality=1 --perf_level=0 --benchmarks=readwhilewriting,readwhilewriting,readwhilewriting --use_existing_db=1 --num=52428800 --threads=1 —writes_per_second=81920 Read Random: From 1.8363 ms/op, improve to 1.7587 ms/op. Read while writing: From 2.985 ms/op, improve to 2.924 ms/op. Test Plan: make all check Reviewers: ljin, haobo, yhchiang, sdong Reviewed By: sdong Subscribers: dhruba, igor Differential Revision: https://reviews.facebook.net/D19419
2014-07-10 07:14:39 +02:00
for (size_t i = 0; i < file_level.num_files; i++) {
const FdWithKeyRange* f = &(file_level.files[i]);
if (AfterFile(ucmp, smallest_user_key, f) ||
BeforeFile(ucmp, largest_user_key, f)) {
// No overlap
} else {
return true; // Overlap
}
}
return false;
}
// Binary search over file list
uint32_t index = 0;
if (smallest_user_key != nullptr) {
// Find the earliest possible internal key for smallest_user_key
InternalKey small(*smallest_user_key, kMaxSequenceNumber,kValueTypeForSeek);
create compressed_levels_ in Version, allocate its space using arena. Make Version::Get, Version::FindFile faster Summary: Define CompressedFileMetaData that just contains fd, smallest_slice, largest_slice. Create compressed_levels_ in Version, the space is allocated using arena Thus increase the file meta data locality, speed up "Get" and "FindFile" benchmark with in-memory tmpfs, could have 4% improvement under "random read" and 2% improvement under "read while writing" benchmark command: ./db_bench --db=/mnt/db/rocksdb --num_levels=6 --key_size=20 --prefix_size=20 --keys_per_prefix=0 --value_size=100 --block_size=4096 --cache_size=17179869184 --cache_numshardbits=6 --compression_type=none --compression_ratio=1 --min_level_to_compress=-1 --disable_seek_compaction=1 --hard_rate_limit=2 --write_buffer_size=134217728 --max_write_buffer_number=2 --level0_file_num_compaction_trigger=8 --target_file_size_base=33554432 --max_bytes_for_level_base=1073741824 --disable_wal=0 --sync=0 --disable_data_sync=1 --verify_checksum=1 --delete_obsolete_files_period_micros=314572800 --max_grandparent_overlap_factor=10 --max_background_compactions=4 --max_background_flushes=0 --level0_slowdown_writes_trigger=16 --level0_stop_writes_trigger=24 --statistics=0 --stats_per_interval=0 --stats_interval=1048576 --histogram=0 --use_plain_table=1 --open_files=-1 --mmap_read=1 --mmap_write=0 --memtablerep=prefix_hash --bloom_bits=10 --bloom_locality=1 --perf_level=0 --benchmarks=readwhilewriting,readwhilewriting,readwhilewriting --use_existing_db=1 --num=52428800 --threads=1 —writes_per_second=81920 Read Random: From 1.8363 ms/op, improve to 1.7587 ms/op. Read while writing: From 2.985 ms/op, improve to 2.924 ms/op. Test Plan: make all check Reviewers: ljin, haobo, yhchiang, sdong Reviewed By: sdong Subscribers: dhruba, igor Differential Revision: https://reviews.facebook.net/D19419
2014-07-10 07:14:39 +02:00
index = FindFile(icmp, file_level, small.Encode());
}
create compressed_levels_ in Version, allocate its space using arena. Make Version::Get, Version::FindFile faster Summary: Define CompressedFileMetaData that just contains fd, smallest_slice, largest_slice. Create compressed_levels_ in Version, the space is allocated using arena Thus increase the file meta data locality, speed up "Get" and "FindFile" benchmark with in-memory tmpfs, could have 4% improvement under "random read" and 2% improvement under "read while writing" benchmark command: ./db_bench --db=/mnt/db/rocksdb --num_levels=6 --key_size=20 --prefix_size=20 --keys_per_prefix=0 --value_size=100 --block_size=4096 --cache_size=17179869184 --cache_numshardbits=6 --compression_type=none --compression_ratio=1 --min_level_to_compress=-1 --disable_seek_compaction=1 --hard_rate_limit=2 --write_buffer_size=134217728 --max_write_buffer_number=2 --level0_file_num_compaction_trigger=8 --target_file_size_base=33554432 --max_bytes_for_level_base=1073741824 --disable_wal=0 --sync=0 --disable_data_sync=1 --verify_checksum=1 --delete_obsolete_files_period_micros=314572800 --max_grandparent_overlap_factor=10 --max_background_compactions=4 --max_background_flushes=0 --level0_slowdown_writes_trigger=16 --level0_stop_writes_trigger=24 --statistics=0 --stats_per_interval=0 --stats_interval=1048576 --histogram=0 --use_plain_table=1 --open_files=-1 --mmap_read=1 --mmap_write=0 --memtablerep=prefix_hash --bloom_bits=10 --bloom_locality=1 --perf_level=0 --benchmarks=readwhilewriting,readwhilewriting,readwhilewriting --use_existing_db=1 --num=52428800 --threads=1 —writes_per_second=81920 Read Random: From 1.8363 ms/op, improve to 1.7587 ms/op. Read while writing: From 2.985 ms/op, improve to 2.924 ms/op. Test Plan: make all check Reviewers: ljin, haobo, yhchiang, sdong Reviewed By: sdong Subscribers: dhruba, igor Differential Revision: https://reviews.facebook.net/D19419
2014-07-10 07:14:39 +02:00
if (index >= file_level.num_files) {
// beginning of range is after all files, so no overlap.
return false;
}
create compressed_levels_ in Version, allocate its space using arena. Make Version::Get, Version::FindFile faster Summary: Define CompressedFileMetaData that just contains fd, smallest_slice, largest_slice. Create compressed_levels_ in Version, the space is allocated using arena Thus increase the file meta data locality, speed up "Get" and "FindFile" benchmark with in-memory tmpfs, could have 4% improvement under "random read" and 2% improvement under "read while writing" benchmark command: ./db_bench --db=/mnt/db/rocksdb --num_levels=6 --key_size=20 --prefix_size=20 --keys_per_prefix=0 --value_size=100 --block_size=4096 --cache_size=17179869184 --cache_numshardbits=6 --compression_type=none --compression_ratio=1 --min_level_to_compress=-1 --disable_seek_compaction=1 --hard_rate_limit=2 --write_buffer_size=134217728 --max_write_buffer_number=2 --level0_file_num_compaction_trigger=8 --target_file_size_base=33554432 --max_bytes_for_level_base=1073741824 --disable_wal=0 --sync=0 --disable_data_sync=1 --verify_checksum=1 --delete_obsolete_files_period_micros=314572800 --max_grandparent_overlap_factor=10 --max_background_compactions=4 --max_background_flushes=0 --level0_slowdown_writes_trigger=16 --level0_stop_writes_trigger=24 --statistics=0 --stats_per_interval=0 --stats_interval=1048576 --histogram=0 --use_plain_table=1 --open_files=-1 --mmap_read=1 --mmap_write=0 --memtablerep=prefix_hash --bloom_bits=10 --bloom_locality=1 --perf_level=0 --benchmarks=readwhilewriting,readwhilewriting,readwhilewriting --use_existing_db=1 --num=52428800 --threads=1 —writes_per_second=81920 Read Random: From 1.8363 ms/op, improve to 1.7587 ms/op. Read while writing: From 2.985 ms/op, improve to 2.924 ms/op. Test Plan: make all check Reviewers: ljin, haobo, yhchiang, sdong Reviewed By: sdong Subscribers: dhruba, igor Differential Revision: https://reviews.facebook.net/D19419
2014-07-10 07:14:39 +02:00
return !BeforeFile(ucmp, largest_user_key, &file_level.files[index]);
}
namespace {
// An internal iterator. For a given version/level pair, yields
// information about the files in the level. For a given entry, key()
// is the largest key that occurs in the file, and value() is an
// 16-byte value containing the file number and file size, both
// encoded using EncodeFixed64.
class LevelFileNumIterator : public Iterator {
public:
LevelFileNumIterator(const InternalKeyComparator& icmp,
const LevelFilesBrief* flevel)
: icmp_(icmp),
flevel_(flevel),
index_(static_cast<uint32_t>(flevel->num_files)),
current_value_(0, 0, 0) { // Marks as invalid
}
virtual bool Valid() const {
return index_ < flevel_->num_files;
}
virtual void Seek(const Slice& target) {
index_ = FindFile(icmp_, *flevel_, target);
}
virtual void SeekToFirst() { index_ = 0; }
virtual void SeekToLast() {
index_ = (flevel_->num_files == 0)
? 0
: static_cast<uint32_t>(flevel_->num_files) - 1;
}
virtual void Next() {
assert(Valid());
index_++;
}
virtual void Prev() {
assert(Valid());
if (index_ == 0) {
index_ = static_cast<uint32_t>(flevel_->num_files); // Marks as invalid
} else {
index_--;
}
}
Slice key() const {
assert(Valid());
return flevel_->files[index_].largest_key;
}
Slice value() const {
assert(Valid());
auto file_meta = flevel_->files[index_];
current_value_ = file_meta.fd;
return Slice(reinterpret_cast<const char*>(&current_value_),
sizeof(FileDescriptor));
}
virtual Status status() const { return Status::OK(); }
private:
const InternalKeyComparator icmp_;
const LevelFilesBrief* flevel_;
uint32_t index_;
mutable FileDescriptor current_value_;
};
class LevelFileIteratorState : public TwoLevelIteratorState {
public:
LevelFileIteratorState(TableCache* table_cache,
const ReadOptions& read_options, const EnvOptions& env_options,
const InternalKeyComparator& icomparator, bool for_compaction,
bool prefix_enabled)
: TwoLevelIteratorState(prefix_enabled),
table_cache_(table_cache), read_options_(read_options),
env_options_(env_options), icomparator_(icomparator),
for_compaction_(for_compaction) {}
Iterator* NewSecondaryIterator(const Slice& meta_handle) override {
if (meta_handle.size() != sizeof(FileDescriptor)) {
return NewErrorIterator(
Status::Corruption("FileReader invoked with unexpected value"));
} else {
const FileDescriptor* fd =
reinterpret_cast<const FileDescriptor*>(meta_handle.data());
return table_cache_->NewIterator(
read_options_, env_options_, icomparator_, *fd,
nullptr /* don't need reference to table*/, for_compaction_);
}
}
bool PrefixMayMatch(const Slice& internal_key) override {
return true;
}
private:
TableCache* table_cache_;
const ReadOptions read_options_;
const EnvOptions& env_options_;
const InternalKeyComparator& icomparator_;
bool for_compaction_;
};
// A wrapper of version builder which references the current version in
// constructor and unref it in the destructor.
// Both of the constructor and destructor need to be called inside DB Mutex.
class BaseReferencedVersionBuilder {
public:
explicit BaseReferencedVersionBuilder(ColumnFamilyData* cfd)
: version_builder_(new VersionBuilder(
cfd->current()->version_set()->env_options(), cfd->table_cache(),
cfd->current()->storage_info())),
version_(cfd->current()) {
version_->Ref();
}
~BaseReferencedVersionBuilder() {
delete version_builder_;
version_->Unref();
}
VersionBuilder* version_builder() { return version_builder_; }
private:
VersionBuilder* version_builder_;
Version* version_;
};
} // anonymous namespace
Status Version::GetTableProperties(std::shared_ptr<const TableProperties>* tp,
const FileMetaData* file_meta,
const std::string* fname) {
auto table_cache = cfd_->table_cache();
auto ioptions = cfd_->ioptions();
Status s = table_cache->GetTableProperties(
vset_->env_options_, cfd_->internal_comparator(), file_meta->fd,
tp, true /* no io */);
if (s.ok()) {
return s;
}
// We only ignore error type `Incomplete` since it's by design that we
// disallow table when it's not in table cache.
if (!s.IsIncomplete()) {
return s;
}
// 2. Table is not present in table cache, we'll read the table properties
// directly from the properties block in the file.
std::unique_ptr<RandomAccessFile> file;
if (fname != nullptr) {
s = ioptions->env->NewRandomAccessFile(
*fname, &file, vset_->env_options_);
} else {
s = ioptions->env->NewRandomAccessFile(
TableFileName(vset_->db_options_->db_paths, file_meta->fd.GetNumber(),
file_meta->fd.GetPathId()),
&file, vset_->env_options_);
}
if (!s.ok()) {
return s;
}
TableProperties* raw_table_properties;
// By setting the magic number to kInvalidTableMagicNumber, we can by
// pass the magic number check in the footer.
s = ReadTableProperties(
file.get(), file_meta->fd.GetFileSize(),
Footer::kInvalidTableMagicNumber /* table's magic number */,
vset_->env_, ioptions->info_log, &raw_table_properties);
if (!s.ok()) {
return s;
}
RecordTick(ioptions->statistics, NUMBER_DIRECT_LOAD_TABLE_PROPERTIES);
*tp = std::shared_ptr<const TableProperties>(raw_table_properties);
return s;
}
Status Version::GetPropertiesOfAllTables(TablePropertiesCollection* props) {
for (int level = 0; level < storage_info_.num_levels_; level++) {
for (const auto& file_meta : storage_info_.files_[level]) {
auto fname =
TableFileName(vset_->db_options_->db_paths, file_meta->fd.GetNumber(),
file_meta->fd.GetPathId());
// 1. If the table is already present in table cache, load table
// properties from there.
std::shared_ptr<const TableProperties> table_properties;
Status s = GetTableProperties(&table_properties, file_meta, &fname);
if (s.ok()) {
props->insert({fname, table_properties});
} else {
return s;
}
}
}
return Status::OK();
}
size_t Version::GetMemoryUsageByTableReaders() {
size_t total_usage = 0;
for (auto& file_level : storage_info_.level_files_brief_) {
for (size_t i = 0; i < file_level.num_files; i++) {
total_usage += cfd_->table_cache()->GetMemoryUsageByTableReader(
vset_->env_options_, cfd_->internal_comparator(),
file_level.files[i].fd);
}
}
return total_usage;
}
void Version::GetColumnFamilyMetaData(ColumnFamilyMetaData* cf_meta) {
assert(cf_meta);
assert(cfd_);
cf_meta->name = cfd_->GetName();
cf_meta->size = 0;
cf_meta->file_count = 0;
cf_meta->levels.clear();
auto* ioptions = cfd_->ioptions();
auto* vstorage = storage_info();
for (int level = 0; level < cfd_->NumberLevels(); level++) {
uint64_t level_size = 0;
cf_meta->file_count += vstorage->LevelFiles(level).size();
std::vector<SstFileMetaData> files;
for (const auto& file : vstorage->LevelFiles(level)) {
uint32_t path_id = file->fd.GetPathId();
std::string file_path;
if (path_id < ioptions->db_paths.size()) {
file_path = ioptions->db_paths[path_id].path;
} else {
assert(!ioptions->db_paths.empty());
file_path = ioptions->db_paths.back().path;
}
files.emplace_back(
MakeTableFileName("", file->fd.GetNumber()),
file_path,
file->fd.GetFileSize(),
file->smallest_seqno,
file->largest_seqno,
file->smallest.user_key().ToString(),
file->largest.user_key().ToString(),
file->being_compacted);
level_size += file->fd.GetFileSize();
}
cf_meta->levels.emplace_back(
level, level_size, std::move(files));
cf_meta->size += level_size;
}
}
uint64_t VersionStorageInfo::GetEstimatedActiveKeys() const {
// Estimation will be not accurate when:
// (1) there is merge keys
// (2) keys are directly overwritten
// (3) deletion on non-existing keys
// (4) low number of samples
if (num_samples_ == 0) {
return 0;
}
uint64_t file_count = 0;
for (int level = 0; level < num_levels_; ++level) {
file_count += files_[level].size();
}
if (num_samples_ < file_count) {
// casting to avoid overflowing
return static_cast<uint64_t>(static_cast<double>(
accumulated_num_non_deletions_ - accumulated_num_deletions_) *
static_cast<double>(file_count) / num_samples_);
} else {
return accumulated_num_non_deletions_ - accumulated_num_deletions_;
}
}
void Version::AddIterators(const ReadOptions& read_options,
const EnvOptions& soptions,
MergeIteratorBuilder* merge_iter_builder) {
assert(storage_info_.finalized_);
if (storage_info_.num_non_empty_levels() == 0) {
// No file in the Version.
return;
}
// Merge all level zero files together since they may overlap
for (size_t i = 0; i < storage_info_.LevelFilesBrief(0).num_files; i++) {
const auto& file = storage_info_.LevelFilesBrief(0).files[i];
merge_iter_builder->AddIterator(cfd_->table_cache()->NewIterator(
read_options, soptions, cfd_->internal_comparator(), file.fd, nullptr,
false, merge_iter_builder->GetArena()));
}
// For levels > 0, we can use a concatenating iterator that sequentially
// walks through the non-overlapping files in the level, opening them
// lazily.
for (int level = 1; level < storage_info_.num_non_empty_levels(); level++) {
if (storage_info_.LevelFilesBrief(level).num_files != 0) {
merge_iter_builder->AddIterator(NewTwoLevelIterator(
new LevelFileIteratorState(
cfd_->table_cache(), read_options, soptions,
cfd_->internal_comparator(), false /* for_compaction */,
cfd_->ioptions()->prefix_extractor != nullptr),
new LevelFileNumIterator(cfd_->internal_comparator(),
&storage_info_.LevelFilesBrief(level)),
merge_iter_builder->GetArena()));
}
}
}
VersionStorageInfo::VersionStorageInfo(
const InternalKeyComparator* internal_comparator,
const Comparator* user_comparator, int levels,
CompactionStyle compaction_style, VersionStorageInfo* ref_vstorage)
: internal_comparator_(internal_comparator),
user_comparator_(user_comparator),
// cfd is nullptr if Version is dummy
num_levels_(levels),
num_non_empty_levels_(0),
file_indexer_(user_comparator),
compaction_style_(compaction_style),
files_(new std::vector<FileMetaData*>[num_levels_]),
files_by_size_(num_levels_),
next_file_to_compact_by_size_(num_levels_),
compaction_score_(num_levels_),
compaction_level_(num_levels_),
accumulated_file_size_(0),
accumulated_raw_key_size_(0),
accumulated_raw_value_size_(0),
accumulated_num_non_deletions_(0),
accumulated_num_deletions_(0),
num_samples_(0),
finalized_(false) {
if (ref_vstorage != nullptr) {
accumulated_file_size_ = ref_vstorage->accumulated_file_size_;
accumulated_raw_key_size_ = ref_vstorage->accumulated_raw_key_size_;
accumulated_raw_value_size_ = ref_vstorage->accumulated_raw_value_size_;
accumulated_num_non_deletions_ =
ref_vstorage->accumulated_num_non_deletions_;
accumulated_num_deletions_ = ref_vstorage->accumulated_num_deletions_;
num_samples_ = ref_vstorage->num_samples_;
}
}
Version::Version(ColumnFamilyData* column_family_data, VersionSet* vset,
uint64_t version_number)
: cfd_(column_family_data),
info_log_((cfd_ == nullptr) ? nullptr : cfd_->ioptions()->info_log),
db_statistics_((cfd_ == nullptr) ? nullptr
: cfd_->ioptions()->statistics),
table_cache_((cfd_ == nullptr) ? nullptr : cfd_->table_cache()),
merge_operator_((cfd_ == nullptr) ? nullptr
: cfd_->ioptions()->merge_operator),
storage_info_((cfd_ == nullptr) ? nullptr : &cfd_->internal_comparator(),
(cfd_ == nullptr) ? nullptr : cfd_->user_comparator(),
cfd_ == nullptr ? 0 : cfd_->NumberLevels(),
cfd_ == nullptr ? kCompactionStyleLevel
: cfd_->ioptions()->compaction_style,
(cfd_ == nullptr || cfd_->current() == nullptr)
? nullptr
: cfd_->current()->storage_info()),
vset_(vset),
next_(this),
prev_(this),
refs_(0),
version_number_(version_number) {}
void Version::Get(const ReadOptions& read_options,
const LookupKey& k,
std::string* value,
[RocksDB] [MergeOperator] The new Merge Interface! Uses merge sequences. Summary: Here are the major changes to the Merge Interface. It has been expanded to handle cases where the MergeOperator is not associative. It does so by stacking up merge operations while scanning through the key history (i.e.: during Get() or Compaction), until a valid Put/Delete/end-of-history is encountered; it then applies all of the merge operations in the correct sequence starting with the base/sentinel value. I have also introduced an "AssociativeMerge" function which allows the user to take advantage of associative merge operations (such as in the case of counters). The implementation will always attempt to merge the operations/operands themselves together when they are encountered, and will resort to the "stacking" method if and only if the "associative-merge" fails. This implementation is conjectured to allow MergeOperator to handle the general case, while still providing the user with the ability to take advantage of certain efficiencies in their own merge-operator / data-structure. NOTE: This is a preliminary diff. This must still go through a lot of review, revision, and testing. Feedback welcome! Test Plan: -This is a preliminary diff. I have only just begun testing/debugging it. -I will be testing this with the existing MergeOperator use-cases and unit-tests (counters, string-append, and redis-lists) -I will be "desk-checking" and walking through the code with the help gdb. -I will find a way of stress-testing the new interface / implementation using db_bench, db_test, merge_test, and/or db_stress. -I will ensure that my tests cover all cases: Get-Memtable, Get-Immutable-Memtable, Get-from-Disk, Iterator-Range-Scan, Flush-Memtable-to-L0, Compaction-L0-L1, Compaction-Ln-L(n+1), Put/Delete found, Put/Delete not-found, end-of-history, end-of-file, etc. -A lot of feedback from the reviewers. Reviewers: haobo, dhruba, zshao, emayanke Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D11499
2013-08-06 05:14:32 +02:00
Status* status,
MergeContext* merge_context,
bool* value_found) {
Slice ikey = k.internal_key();
Slice user_key = k.user_key();
assert(status->ok() || status->IsMergeInProgress());
GetContext get_context(
user_comparator(), merge_operator_, info_log_, db_statistics_,
status->ok() ? GetContext::kNotFound : GetContext::kMerge, user_key,
value, value_found, merge_context);
FilePicker fp(
storage_info_.files_, user_key, ikey, &storage_info_.level_files_brief_,
storage_info_.num_non_empty_levels_, &storage_info_.file_indexer_,
user_comparator(), internal_comparator());
FdWithKeyRange* f = fp.GetNextFile();
while (f != nullptr) {
*status = table_cache_->Get(read_options, *internal_comparator(), f->fd,
ikey, &get_context);
// TODO: examine the behavior for corrupted key
if (!status->ok()) {
return;
}
switch (get_context.State()) {
case GetContext::kNotFound:
// Keep searching in other files
break;
case GetContext::kFound:
return;
case GetContext::kDeleted:
// Use empty error message for speed
*status = Status::NotFound();
return;
case GetContext::kCorrupt:
*status = Status::Corruption("corrupted key for ", user_key);
return;
case GetContext::kMerge:
break;
}
f = fp.GetNextFile();
}
if (GetContext::kMerge == get_context.State()) {
if (!merge_operator_) {
*status = Status::InvalidArgument(
"merge_operator is not properly initialized.");
return;
}
[RocksDB] [MergeOperator] The new Merge Interface! Uses merge sequences. Summary: Here are the major changes to the Merge Interface. It has been expanded to handle cases where the MergeOperator is not associative. It does so by stacking up merge operations while scanning through the key history (i.e.: during Get() or Compaction), until a valid Put/Delete/end-of-history is encountered; it then applies all of the merge operations in the correct sequence starting with the base/sentinel value. I have also introduced an "AssociativeMerge" function which allows the user to take advantage of associative merge operations (such as in the case of counters). The implementation will always attempt to merge the operations/operands themselves together when they are encountered, and will resort to the "stacking" method if and only if the "associative-merge" fails. This implementation is conjectured to allow MergeOperator to handle the general case, while still providing the user with the ability to take advantage of certain efficiencies in their own merge-operator / data-structure. NOTE: This is a preliminary diff. This must still go through a lot of review, revision, and testing. Feedback welcome! Test Plan: -This is a preliminary diff. I have only just begun testing/debugging it. -I will be testing this with the existing MergeOperator use-cases and unit-tests (counters, string-append, and redis-lists) -I will be "desk-checking" and walking through the code with the help gdb. -I will find a way of stress-testing the new interface / implementation using db_bench, db_test, merge_test, and/or db_stress. -I will ensure that my tests cover all cases: Get-Memtable, Get-Immutable-Memtable, Get-from-Disk, Iterator-Range-Scan, Flush-Memtable-to-L0, Compaction-L0-L1, Compaction-Ln-L(n+1), Put/Delete found, Put/Delete not-found, end-of-history, end-of-file, etc. -A lot of feedback from the reviewers. Reviewers: haobo, dhruba, zshao, emayanke Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D11499
2013-08-06 05:14:32 +02:00
// merge_operands are in saver and we hit the beginning of the key history
// do a final merge of nullptr and operands;
if (merge_operator_->FullMerge(user_key, nullptr,
merge_context->GetOperands(), value,
info_log_)) {
[RocksDB] [MergeOperator] The new Merge Interface! Uses merge sequences. Summary: Here are the major changes to the Merge Interface. It has been expanded to handle cases where the MergeOperator is not associative. It does so by stacking up merge operations while scanning through the key history (i.e.: during Get() or Compaction), until a valid Put/Delete/end-of-history is encountered; it then applies all of the merge operations in the correct sequence starting with the base/sentinel value. I have also introduced an "AssociativeMerge" function which allows the user to take advantage of associative merge operations (such as in the case of counters). The implementation will always attempt to merge the operations/operands themselves together when they are encountered, and will resort to the "stacking" method if and only if the "associative-merge" fails. This implementation is conjectured to allow MergeOperator to handle the general case, while still providing the user with the ability to take advantage of certain efficiencies in their own merge-operator / data-structure. NOTE: This is a preliminary diff. This must still go through a lot of review, revision, and testing. Feedback welcome! Test Plan: -This is a preliminary diff. I have only just begun testing/debugging it. -I will be testing this with the existing MergeOperator use-cases and unit-tests (counters, string-append, and redis-lists) -I will be "desk-checking" and walking through the code with the help gdb. -I will find a way of stress-testing the new interface / implementation using db_bench, db_test, merge_test, and/or db_stress. -I will ensure that my tests cover all cases: Get-Memtable, Get-Immutable-Memtable, Get-from-Disk, Iterator-Range-Scan, Flush-Memtable-to-L0, Compaction-L0-L1, Compaction-Ln-L(n+1), Put/Delete found, Put/Delete not-found, end-of-history, end-of-file, etc. -A lot of feedback from the reviewers. Reviewers: haobo, dhruba, zshao, emayanke Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D11499
2013-08-06 05:14:32 +02:00
*status = Status::OK();
} else {
RecordTick(db_statistics_, NUMBER_MERGE_FAILURES);
[RocksDB] [MergeOperator] The new Merge Interface! Uses merge sequences. Summary: Here are the major changes to the Merge Interface. It has been expanded to handle cases where the MergeOperator is not associative. It does so by stacking up merge operations while scanning through the key history (i.e.: during Get() or Compaction), until a valid Put/Delete/end-of-history is encountered; it then applies all of the merge operations in the correct sequence starting with the base/sentinel value. I have also introduced an "AssociativeMerge" function which allows the user to take advantage of associative merge operations (such as in the case of counters). The implementation will always attempt to merge the operations/operands themselves together when they are encountered, and will resort to the "stacking" method if and only if the "associative-merge" fails. This implementation is conjectured to allow MergeOperator to handle the general case, while still providing the user with the ability to take advantage of certain efficiencies in their own merge-operator / data-structure. NOTE: This is a preliminary diff. This must still go through a lot of review, revision, and testing. Feedback welcome! Test Plan: -This is a preliminary diff. I have only just begun testing/debugging it. -I will be testing this with the existing MergeOperator use-cases and unit-tests (counters, string-append, and redis-lists) -I will be "desk-checking" and walking through the code with the help gdb. -I will find a way of stress-testing the new interface / implementation using db_bench, db_test, merge_test, and/or db_stress. -I will ensure that my tests cover all cases: Get-Memtable, Get-Immutable-Memtable, Get-from-Disk, Iterator-Range-Scan, Flush-Memtable-to-L0, Compaction-L0-L1, Compaction-Ln-L(n+1), Put/Delete found, Put/Delete not-found, end-of-history, end-of-file, etc. -A lot of feedback from the reviewers. Reviewers: haobo, dhruba, zshao, emayanke Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D11499
2013-08-06 05:14:32 +02:00
*status = Status::Corruption("could not perform end-of-key merge for ",
user_key);
}
} else {
*status = Status::NotFound(); // Use an empty error message for speed
}
}
void VersionStorageInfo::GenerateLevelFilesBrief() {
level_files_brief_.resize(num_non_empty_levels_);
create compressed_levels_ in Version, allocate its space using arena. Make Version::Get, Version::FindFile faster Summary: Define CompressedFileMetaData that just contains fd, smallest_slice, largest_slice. Create compressed_levels_ in Version, the space is allocated using arena Thus increase the file meta data locality, speed up "Get" and "FindFile" benchmark with in-memory tmpfs, could have 4% improvement under "random read" and 2% improvement under "read while writing" benchmark command: ./db_bench --db=/mnt/db/rocksdb --num_levels=6 --key_size=20 --prefix_size=20 --keys_per_prefix=0 --value_size=100 --block_size=4096 --cache_size=17179869184 --cache_numshardbits=6 --compression_type=none --compression_ratio=1 --min_level_to_compress=-1 --disable_seek_compaction=1 --hard_rate_limit=2 --write_buffer_size=134217728 --max_write_buffer_number=2 --level0_file_num_compaction_trigger=8 --target_file_size_base=33554432 --max_bytes_for_level_base=1073741824 --disable_wal=0 --sync=0 --disable_data_sync=1 --verify_checksum=1 --delete_obsolete_files_period_micros=314572800 --max_grandparent_overlap_factor=10 --max_background_compactions=4 --max_background_flushes=0 --level0_slowdown_writes_trigger=16 --level0_stop_writes_trigger=24 --statistics=0 --stats_per_interval=0 --stats_interval=1048576 --histogram=0 --use_plain_table=1 --open_files=-1 --mmap_read=1 --mmap_write=0 --memtablerep=prefix_hash --bloom_bits=10 --bloom_locality=1 --perf_level=0 --benchmarks=readwhilewriting,readwhilewriting,readwhilewriting --use_existing_db=1 --num=52428800 --threads=1 —writes_per_second=81920 Read Random: From 1.8363 ms/op, improve to 1.7587 ms/op. Read while writing: From 2.985 ms/op, improve to 2.924 ms/op. Test Plan: make all check Reviewers: ljin, haobo, yhchiang, sdong Reviewed By: sdong Subscribers: dhruba, igor Differential Revision: https://reviews.facebook.net/D19419
2014-07-10 07:14:39 +02:00
for (int level = 0; level < num_non_empty_levels_; level++) {
DoGenerateLevelFilesBrief(
&level_files_brief_[level], files_[level], &arena_);
create compressed_levels_ in Version, allocate its space using arena. Make Version::Get, Version::FindFile faster Summary: Define CompressedFileMetaData that just contains fd, smallest_slice, largest_slice. Create compressed_levels_ in Version, the space is allocated using arena Thus increase the file meta data locality, speed up "Get" and "FindFile" benchmark with in-memory tmpfs, could have 4% improvement under "random read" and 2% improvement under "read while writing" benchmark command: ./db_bench --db=/mnt/db/rocksdb --num_levels=6 --key_size=20 --prefix_size=20 --keys_per_prefix=0 --value_size=100 --block_size=4096 --cache_size=17179869184 --cache_numshardbits=6 --compression_type=none --compression_ratio=1 --min_level_to_compress=-1 --disable_seek_compaction=1 --hard_rate_limit=2 --write_buffer_size=134217728 --max_write_buffer_number=2 --level0_file_num_compaction_trigger=8 --target_file_size_base=33554432 --max_bytes_for_level_base=1073741824 --disable_wal=0 --sync=0 --disable_data_sync=1 --verify_checksum=1 --delete_obsolete_files_period_micros=314572800 --max_grandparent_overlap_factor=10 --max_background_compactions=4 --max_background_flushes=0 --level0_slowdown_writes_trigger=16 --level0_stop_writes_trigger=24 --statistics=0 --stats_per_interval=0 --stats_interval=1048576 --histogram=0 --use_plain_table=1 --open_files=-1 --mmap_read=1 --mmap_write=0 --memtablerep=prefix_hash --bloom_bits=10 --bloom_locality=1 --perf_level=0 --benchmarks=readwhilewriting,readwhilewriting,readwhilewriting --use_existing_db=1 --num=52428800 --threads=1 —writes_per_second=81920 Read Random: From 1.8363 ms/op, improve to 1.7587 ms/op. Read while writing: From 2.985 ms/op, improve to 2.924 ms/op. Test Plan: make all check Reviewers: ljin, haobo, yhchiang, sdong Reviewed By: sdong Subscribers: dhruba, igor Differential Revision: https://reviews.facebook.net/D19419
2014-07-10 07:14:39 +02:00
}
}
void Version::PrepareApply(const MutableCFOptions& mutable_cf_options,
std::vector<uint64_t>& size_being_compacted) {
UpdateAccumulatedStats();
storage_info_.ComputeCompactionScore(
mutable_cf_options, cfd_->ioptions()->compaction_options_fifo,
size_being_compacted);
storage_info_.UpdateFilesBySize();
storage_info_.UpdateNumNonEmptyLevels();
storage_info_.GenerateFileIndexer();
storage_info_.GenerateLevelFilesBrief();
}
bool Version::MaybeInitializeFileMetaData(FileMetaData* file_meta) {
if (file_meta->init_stats_from_file ||
file_meta->compensated_file_size > 0) {
return false;
}
std::shared_ptr<const TableProperties> tp;
Status s = GetTableProperties(&tp, file_meta);
file_meta->init_stats_from_file = true;
if (!s.ok()) {
Log(InfoLogLevel::ERROR_LEVEL, vset_->db_options_->info_log,
"Unable to load table properties for file %" PRIu64 " --- %s\n",
file_meta->fd.GetNumber(), s.ToString().c_str());
return false;
}
if (tp.get() == nullptr) return false;
file_meta->num_entries = tp->num_entries;
file_meta->num_deletions = GetDeletedKeys(tp->user_collected_properties);
file_meta->raw_value_size = tp->raw_value_size;
file_meta->raw_key_size = tp->raw_key_size;
return true;
}
void VersionStorageInfo::UpdateAccumulatedStats(FileMetaData* file_meta) {
assert(file_meta->init_stats_from_file);
accumulated_file_size_ += file_meta->fd.GetFileSize();
accumulated_raw_key_size_ += file_meta->raw_key_size;
accumulated_raw_value_size_ += file_meta->raw_value_size;
accumulated_num_non_deletions_ +=
file_meta->num_entries - file_meta->num_deletions;
accumulated_num_deletions_ += file_meta->num_deletions;
num_samples_++;
}
void Version::UpdateAccumulatedStats() {
// maximum number of table properties loaded from files.
const int kMaxInitCount = 20;
int init_count = 0;
// here only the first kMaxInitCount files which haven't been
// initialized from file will be updated with num_deletions.
// The motivation here is to cap the maximum I/O per Version creation.
// The reason for choosing files from lower-level instead of higher-level
// is that such design is able to propagate the initialization from
// lower-level to higher-level: When the num_deletions of lower-level
// files are updated, it will make the lower-level files have accurate
// compensated_file_size, making lower-level to higher-level compaction
// will be triggered, which creates higher-level files whose num_deletions
// will be updated here.
for (int level = 0;
level < storage_info_.num_levels_ && init_count < kMaxInitCount;
++level) {
for (auto* file_meta : storage_info_.files_[level]) {
if (MaybeInitializeFileMetaData(file_meta)) {
// each FileMeta will be initialized only once.
storage_info_.UpdateAccumulatedStats(file_meta);
if (++init_count >= kMaxInitCount) {
break;
}
}
}
}
// In case all sampled-files contain only deletion entries, then we
// load the table-property of a file in higher-level to initialize
// that value.
for (int level = storage_info_.num_levels_ - 1;
storage_info_.accumulated_raw_value_size_ == 0 && level >= 0; --level) {
for (int i = static_cast<int>(storage_info_.files_[level].size()) - 1;
storage_info_.accumulated_raw_value_size_ == 0 && i >= 0; --i) {
if (MaybeInitializeFileMetaData(storage_info_.files_[level][i])) {
storage_info_.UpdateAccumulatedStats(storage_info_.files_[level][i]);
}
}
}
storage_info_.ComputeCompensatedSizes();
}
void VersionStorageInfo::ComputeCompensatedSizes() {
static const int kDeletionWeightOnCompaction = 2;
uint64_t average_value_size = GetAverageValueSize();
// compute the compensated size
for (int level = 0; level < num_levels_; level++) {
for (auto* file_meta : files_[level]) {
// Here we only compute compensated_file_size for those file_meta
// which compensated_file_size is uninitialized (== 0).
if (file_meta->compensated_file_size == 0) {
file_meta->compensated_file_size = file_meta->fd.GetFileSize() +
file_meta->num_deletions * average_value_size *
kDeletionWeightOnCompaction;
}
}
}
}
int VersionStorageInfo::MaxInputLevel() const {
if (compaction_style_ == kCompactionStyleLevel) {
return num_levels() - 2;
}
return 0;
}
void VersionStorageInfo::ComputeCompactionScore(
const MutableCFOptions& mutable_cf_options,
const CompactionOptionsFIFO& compaction_options_fifo,
std::vector<uint64_t>& size_being_compacted) {
double max_score = 0;
int max_score_level = 0;
for (int level = 0; level <= MaxInputLevel(); level++) {
double score;
if (level == 0) {
// We treat level-0 specially by bounding the number of files
// instead of number of bytes for two reasons:
//
// (1) With larger write-buffer sizes, it is nice not to do too
// many level-0 compactions.
//
// (2) The files in level-0 are merged on every read and
// therefore we wish to avoid too many files when the individual
// file size is small (perhaps because of a small write-buffer
// setting, or very high compression ratios, or lots of
// overwrites/deletions).
int numfiles = 0;
uint64_t total_size = 0;
for (unsigned int i = 0; i < files_[level].size(); i++) {
if (!files_[level][i]->being_compacted) {
total_size += files_[level][i]->compensated_file_size;
numfiles++;
}
}
if (compaction_style_ == kCompactionStyleFIFO) {
score = static_cast<double>(total_size) /
compaction_options_fifo.max_table_files_size;
} else if (numfiles >= mutable_cf_options.level0_stop_writes_trigger) {
// If we are slowing down writes, then we better compact that first
score = 1000000;
} else if (numfiles >=
mutable_cf_options.level0_slowdown_writes_trigger) {
score = 10000;
} else {
score = static_cast<double>(numfiles) /
mutable_cf_options.level0_file_num_compaction_trigger;
}
} else {
// Compute the ratio of current size to size limit.
const uint64_t level_bytes =
TotalCompensatedFileSize(files_[level]) - size_being_compacted[level];
score = static_cast<double>(level_bytes) /
mutable_cf_options.MaxBytesForLevel(level);
if (max_score < score) {
max_score = score;
max_score_level = level;
}
}
compaction_level_[level] = level;
compaction_score_[level] = score;
}
// update the max compaction score in levels 1 to n-1
max_compaction_score_ = max_score;
max_compaction_score_level_ = max_score_level;
// sort all the levels based on their score. Higher scores get listed
// first. Use bubble sort because the number of entries are small.
for (int i = 0; i < num_levels() - 2; i++) {
for (int j = i + 1; j < num_levels() - 1; j++) {
if (compaction_score_[i] < compaction_score_[j]) {
double score = compaction_score_[i];
int level = compaction_level_[i];
compaction_score_[i] = compaction_score_[j];
compaction_level_[i] = compaction_level_[j];
compaction_score_[j] = score;
compaction_level_[j] = level;
}
}
}
}
namespace {
// used to sort files by size
struct Fsize {
int index;
FileMetaData* file;
};
// Compator that is used to sort files based on their size
// In normal mode: descending size
bool CompareCompensatedSizeDescending(const Fsize& first, const Fsize& second) {
return (first.file->compensated_file_size >
second.file->compensated_file_size);
}
} // anonymous namespace
void VersionStorageInfo::AddFile(int level, FileMetaData* f) {
assert(level < num_levels());
auto* level_files = &files_[level];
// Must not overlap
assert(level <= 0 || level_files->empty() ||
internal_comparator_->Compare(
(*level_files)[level_files->size() - 1]->largest, f->smallest) <
0);
f->refs++;
level_files->push_back(f);
}
void VersionStorageInfo::UpdateNumNonEmptyLevels() {
num_non_empty_levels_ = num_levels_;
for (int i = num_levels_ - 1; i >= 0; i--) {
if (files_[i].size() != 0) {
return;
} else {
num_non_empty_levels_ = i;
}
}
}
void VersionStorageInfo::UpdateFilesBySize() {
if (compaction_style_ == kCompactionStyleFIFO ||
compaction_style_ == kCompactionStyleUniversal) {
// don't need this
return;
}
// No need to sort the highest level because it is never compacted.
for (int level = 0; level < num_levels() - 1; level++) {
const std::vector<FileMetaData*>& files = files_[level];
auto& files_by_size = files_by_size_[level];
assert(files_by_size.size() == 0);
// populate a temp vector for sorting based on size
std::vector<Fsize> temp(files.size());
for (unsigned int i = 0; i < files.size(); i++) {
temp[i].index = i;
temp[i].file = files[i];
}
// sort the top number_of_files_to_sort_ based on file size
size_t num = VersionStorageInfo::kNumberFilesToSort;
if (num > temp.size()) {
num = temp.size();
}
std::partial_sort(temp.begin(), temp.begin() + num, temp.end(),
CompareCompensatedSizeDescending);
assert(temp.size() == files.size());
// initialize files_by_size_
for (unsigned int i = 0; i < temp.size(); i++) {
files_by_size.push_back(temp[i].index);
}
next_file_to_compact_by_size_[level] = 0;
assert(files_[level].size() == files_by_size_[level].size());
}
}
void Version::Ref() {
++refs_;
}
bool Version::Unref() {
assert(refs_ >= 1);
--refs_;
if (refs_ == 0) {
delete this;
return true;
}
return false;
}
bool VersionStorageInfo::OverlapInLevel(int level,
const Slice* smallest_user_key,
const Slice* largest_user_key) {
return SomeFileOverlapsRange(*internal_comparator_, (level > 0),
level_files_brief_[level], smallest_user_key,
largest_user_key);
}
int VersionStorageInfo::PickLevelForMemTableOutput(
const MutableCFOptions& mutable_cf_options, const Slice& smallest_user_key,
const Slice& largest_user_key) {
int level = 0;
if (!OverlapInLevel(0, &smallest_user_key, &largest_user_key)) {
// Push to next level if there is no overlap in next level,
// and the #bytes overlapping in the level after that are limited.
InternalKey start(smallest_user_key, kMaxSequenceNumber, kValueTypeForSeek);
InternalKey limit(largest_user_key, 0, static_cast<ValueType>(0));
std::vector<FileMetaData*> overlaps;
while (mutable_cf_options.max_mem_compaction_level > 0 &&
level < mutable_cf_options.max_mem_compaction_level) {
if (OverlapInLevel(level + 1, &smallest_user_key, &largest_user_key)) {
break;
}
if (level + 2 >= num_levels_) {
level++;
break;
}
GetOverlappingInputs(level + 2, &start, &limit, &overlaps);
const uint64_t sum = TotalFileSize(overlaps);
if (sum > mutable_cf_options.MaxGrandParentOverlapBytes(level)) {
break;
}
level++;
}
}
return level;
}
// Store in "*inputs" all files in "level" that overlap [begin,end]
// If hint_index is specified, then it points to a file in the
// overlapping range.
// The file_index returns a pointer to any file in an overlapping range.
void VersionStorageInfo::GetOverlappingInputs(
int level, const InternalKey* begin, const InternalKey* end,
std::vector<FileMetaData*>* inputs, int hint_index, int* file_index) {
inputs->clear();
Slice user_begin, user_end;
if (begin != nullptr) {
user_begin = begin->user_key();
}
if (end != nullptr) {
user_end = end->user_key();
}
if (file_index) {
*file_index = -1;
}
const Comparator* user_cmp = user_comparator_;
if (begin != nullptr && end != nullptr && level > 0) {
GetOverlappingInputsBinarySearch(level, user_begin, user_end, inputs,
hint_index, file_index);
return;
}
for (size_t i = 0; i < level_files_brief_[level].num_files; ) {
FdWithKeyRange* f = &(level_files_brief_[level].files[i++]);
create compressed_levels_ in Version, allocate its space using arena. Make Version::Get, Version::FindFile faster Summary: Define CompressedFileMetaData that just contains fd, smallest_slice, largest_slice. Create compressed_levels_ in Version, the space is allocated using arena Thus increase the file meta data locality, speed up "Get" and "FindFile" benchmark with in-memory tmpfs, could have 4% improvement under "random read" and 2% improvement under "read while writing" benchmark command: ./db_bench --db=/mnt/db/rocksdb --num_levels=6 --key_size=20 --prefix_size=20 --keys_per_prefix=0 --value_size=100 --block_size=4096 --cache_size=17179869184 --cache_numshardbits=6 --compression_type=none --compression_ratio=1 --min_level_to_compress=-1 --disable_seek_compaction=1 --hard_rate_limit=2 --write_buffer_size=134217728 --max_write_buffer_number=2 --level0_file_num_compaction_trigger=8 --target_file_size_base=33554432 --max_bytes_for_level_base=1073741824 --disable_wal=0 --sync=0 --disable_data_sync=1 --verify_checksum=1 --delete_obsolete_files_period_micros=314572800 --max_grandparent_overlap_factor=10 --max_background_compactions=4 --max_background_flushes=0 --level0_slowdown_writes_trigger=16 --level0_stop_writes_trigger=24 --statistics=0 --stats_per_interval=0 --stats_interval=1048576 --histogram=0 --use_plain_table=1 --open_files=-1 --mmap_read=1 --mmap_write=0 --memtablerep=prefix_hash --bloom_bits=10 --bloom_locality=1 --perf_level=0 --benchmarks=readwhilewriting,readwhilewriting,readwhilewriting --use_existing_db=1 --num=52428800 --threads=1 —writes_per_second=81920 Read Random: From 1.8363 ms/op, improve to 1.7587 ms/op. Read while writing: From 2.985 ms/op, improve to 2.924 ms/op. Test Plan: make all check Reviewers: ljin, haobo, yhchiang, sdong Reviewed By: sdong Subscribers: dhruba, igor Differential Revision: https://reviews.facebook.net/D19419
2014-07-10 07:14:39 +02:00
const Slice file_start = ExtractUserKey(f->smallest_key);
const Slice file_limit = ExtractUserKey(f->largest_key);
if (begin != nullptr && user_cmp->Compare(file_limit, user_begin) < 0) {
// "f" is completely before specified range; skip it
} else if (end != nullptr && user_cmp->Compare(file_start, user_end) > 0) {
// "f" is completely after specified range; skip it
} else {
create compressed_levels_ in Version, allocate its space using arena. Make Version::Get, Version::FindFile faster Summary: Define CompressedFileMetaData that just contains fd, smallest_slice, largest_slice. Create compressed_levels_ in Version, the space is allocated using arena Thus increase the file meta data locality, speed up "Get" and "FindFile" benchmark with in-memory tmpfs, could have 4% improvement under "random read" and 2% improvement under "read while writing" benchmark command: ./db_bench --db=/mnt/db/rocksdb --num_levels=6 --key_size=20 --prefix_size=20 --keys_per_prefix=0 --value_size=100 --block_size=4096 --cache_size=17179869184 --cache_numshardbits=6 --compression_type=none --compression_ratio=1 --min_level_to_compress=-1 --disable_seek_compaction=1 --hard_rate_limit=2 --write_buffer_size=134217728 --max_write_buffer_number=2 --level0_file_num_compaction_trigger=8 --target_file_size_base=33554432 --max_bytes_for_level_base=1073741824 --disable_wal=0 --sync=0 --disable_data_sync=1 --verify_checksum=1 --delete_obsolete_files_period_micros=314572800 --max_grandparent_overlap_factor=10 --max_background_compactions=4 --max_background_flushes=0 --level0_slowdown_writes_trigger=16 --level0_stop_writes_trigger=24 --statistics=0 --stats_per_interval=0 --stats_interval=1048576 --histogram=0 --use_plain_table=1 --open_files=-1 --mmap_read=1 --mmap_write=0 --memtablerep=prefix_hash --bloom_bits=10 --bloom_locality=1 --perf_level=0 --benchmarks=readwhilewriting,readwhilewriting,readwhilewriting --use_existing_db=1 --num=52428800 --threads=1 —writes_per_second=81920 Read Random: From 1.8363 ms/op, improve to 1.7587 ms/op. Read while writing: From 2.985 ms/op, improve to 2.924 ms/op. Test Plan: make all check Reviewers: ljin, haobo, yhchiang, sdong Reviewed By: sdong Subscribers: dhruba, igor Differential Revision: https://reviews.facebook.net/D19419
2014-07-10 07:14:39 +02:00
inputs->push_back(files_[level][i-1]);
if (level == 0) {
// Level-0 files may overlap each other. So check if the newly
// added file has expanded the range. If so, restart search.
if (begin != nullptr && user_cmp->Compare(file_start, user_begin) < 0) {
user_begin = file_start;
inputs->clear();
i = 0;
} else if (end != nullptr
&& user_cmp->Compare(file_limit, user_end) > 0) {
user_end = file_limit;
inputs->clear();
i = 0;
}
} else if (file_index) {
*file_index = static_cast<int>(i) - 1;
}
}
}
}
// Store in "*inputs" all files in "level" that overlap [begin,end]
// Employ binary search to find at least one file that overlaps the
// specified range. From that file, iterate backwards and
// forwards to find all overlapping files.
void VersionStorageInfo::GetOverlappingInputsBinarySearch(
int level, const Slice& user_begin, const Slice& user_end,
std::vector<FileMetaData*>* inputs, int hint_index, int* file_index) {
assert(level > 0);
int min = 0;
int mid = 0;
int max = static_cast<int>(files_[level].size()) - 1;
bool foundOverlap = false;
const Comparator* user_cmp = user_comparator_;
// if the caller already knows the index of a file that has overlap,
// then we can skip the binary search.
if (hint_index != -1) {
mid = hint_index;
foundOverlap = true;
}
while (!foundOverlap && min <= max) {
mid = (min + max)/2;
FdWithKeyRange* f = &(level_files_brief_[level].files[mid]);
create compressed_levels_ in Version, allocate its space using arena. Make Version::Get, Version::FindFile faster Summary: Define CompressedFileMetaData that just contains fd, smallest_slice, largest_slice. Create compressed_levels_ in Version, the space is allocated using arena Thus increase the file meta data locality, speed up "Get" and "FindFile" benchmark with in-memory tmpfs, could have 4% improvement under "random read" and 2% improvement under "read while writing" benchmark command: ./db_bench --db=/mnt/db/rocksdb --num_levels=6 --key_size=20 --prefix_size=20 --keys_per_prefix=0 --value_size=100 --block_size=4096 --cache_size=17179869184 --cache_numshardbits=6 --compression_type=none --compression_ratio=1 --min_level_to_compress=-1 --disable_seek_compaction=1 --hard_rate_limit=2 --write_buffer_size=134217728 --max_write_buffer_number=2 --level0_file_num_compaction_trigger=8 --target_file_size_base=33554432 --max_bytes_for_level_base=1073741824 --disable_wal=0 --sync=0 --disable_data_sync=1 --verify_checksum=1 --delete_obsolete_files_period_micros=314572800 --max_grandparent_overlap_factor=10 --max_background_compactions=4 --max_background_flushes=0 --level0_slowdown_writes_trigger=16 --level0_stop_writes_trigger=24 --statistics=0 --stats_per_interval=0 --stats_interval=1048576 --histogram=0 --use_plain_table=1 --open_files=-1 --mmap_read=1 --mmap_write=0 --memtablerep=prefix_hash --bloom_bits=10 --bloom_locality=1 --perf_level=0 --benchmarks=readwhilewriting,readwhilewriting,readwhilewriting --use_existing_db=1 --num=52428800 --threads=1 —writes_per_second=81920 Read Random: From 1.8363 ms/op, improve to 1.7587 ms/op. Read while writing: From 2.985 ms/op, improve to 2.924 ms/op. Test Plan: make all check Reviewers: ljin, haobo, yhchiang, sdong Reviewed By: sdong Subscribers: dhruba, igor Differential Revision: https://reviews.facebook.net/D19419
2014-07-10 07:14:39 +02:00
const Slice file_start = ExtractUserKey(f->smallest_key);
const Slice file_limit = ExtractUserKey(f->largest_key);
if (user_cmp->Compare(file_limit, user_begin) < 0) {
min = mid + 1;
} else if (user_cmp->Compare(user_end, file_start) < 0) {
max = mid - 1;
} else {
foundOverlap = true;
break;
}
}
// If there were no overlapping files, return immediately.
if (!foundOverlap) {
return;
}
// returns the index where an overlap is found
if (file_index) {
*file_index = mid;
}
ExtendOverlappingInputs(level, user_begin, user_end, inputs, mid);
}
// Store in "*inputs" all files in "level" that overlap [begin,end]
// The midIndex specifies the index of at least one file that
// overlaps the specified range. From that file, iterate backward
// and forward to find all overlapping files.
// Use FileLevel in searching, make it faster
void VersionStorageInfo::ExtendOverlappingInputs(
int level, const Slice& user_begin, const Slice& user_end,
std::vector<FileMetaData*>* inputs, unsigned int midIndex) {
const Comparator* user_cmp = user_comparator_;
const FdWithKeyRange* files = level_files_brief_[level].files;
#ifndef NDEBUG
{
// assert that the file at midIndex overlaps with the range
assert(midIndex < level_files_brief_[level].num_files);
create compressed_levels_ in Version, allocate its space using arena. Make Version::Get, Version::FindFile faster Summary: Define CompressedFileMetaData that just contains fd, smallest_slice, largest_slice. Create compressed_levels_ in Version, the space is allocated using arena Thus increase the file meta data locality, speed up "Get" and "FindFile" benchmark with in-memory tmpfs, could have 4% improvement under "random read" and 2% improvement under "read while writing" benchmark command: ./db_bench --db=/mnt/db/rocksdb --num_levels=6 --key_size=20 --prefix_size=20 --keys_per_prefix=0 --value_size=100 --block_size=4096 --cache_size=17179869184 --cache_numshardbits=6 --compression_type=none --compression_ratio=1 --min_level_to_compress=-1 --disable_seek_compaction=1 --hard_rate_limit=2 --write_buffer_size=134217728 --max_write_buffer_number=2 --level0_file_num_compaction_trigger=8 --target_file_size_base=33554432 --max_bytes_for_level_base=1073741824 --disable_wal=0 --sync=0 --disable_data_sync=1 --verify_checksum=1 --delete_obsolete_files_period_micros=314572800 --max_grandparent_overlap_factor=10 --max_background_compactions=4 --max_background_flushes=0 --level0_slowdown_writes_trigger=16 --level0_stop_writes_trigger=24 --statistics=0 --stats_per_interval=0 --stats_interval=1048576 --histogram=0 --use_plain_table=1 --open_files=-1 --mmap_read=1 --mmap_write=0 --memtablerep=prefix_hash --bloom_bits=10 --bloom_locality=1 --perf_level=0 --benchmarks=readwhilewriting,readwhilewriting,readwhilewriting --use_existing_db=1 --num=52428800 --threads=1 —writes_per_second=81920 Read Random: From 1.8363 ms/op, improve to 1.7587 ms/op. Read while writing: From 2.985 ms/op, improve to 2.924 ms/op. Test Plan: make all check Reviewers: ljin, haobo, yhchiang, sdong Reviewed By: sdong Subscribers: dhruba, igor Differential Revision: https://reviews.facebook.net/D19419
2014-07-10 07:14:39 +02:00
const FdWithKeyRange* f = &files[midIndex];
const Slice fstart = ExtractUserKey(f->smallest_key);
const Slice flimit = ExtractUserKey(f->largest_key);
if (user_cmp->Compare(fstart, user_begin) >= 0) {
assert(user_cmp->Compare(fstart, user_end) <= 0);
} else {
assert(user_cmp->Compare(flimit, user_begin) >= 0);
}
}
#endif
int startIndex = midIndex + 1;
int endIndex = midIndex;
int count __attribute__((unused)) = 0;
// check backwards from 'mid' to lower indices
for (int i = midIndex; i >= 0 ; i--) {
create compressed_levels_ in Version, allocate its space using arena. Make Version::Get, Version::FindFile faster Summary: Define CompressedFileMetaData that just contains fd, smallest_slice, largest_slice. Create compressed_levels_ in Version, the space is allocated using arena Thus increase the file meta data locality, speed up "Get" and "FindFile" benchmark with in-memory tmpfs, could have 4% improvement under "random read" and 2% improvement under "read while writing" benchmark command: ./db_bench --db=/mnt/db/rocksdb --num_levels=6 --key_size=20 --prefix_size=20 --keys_per_prefix=0 --value_size=100 --block_size=4096 --cache_size=17179869184 --cache_numshardbits=6 --compression_type=none --compression_ratio=1 --min_level_to_compress=-1 --disable_seek_compaction=1 --hard_rate_limit=2 --write_buffer_size=134217728 --max_write_buffer_number=2 --level0_file_num_compaction_trigger=8 --target_file_size_base=33554432 --max_bytes_for_level_base=1073741824 --disable_wal=0 --sync=0 --disable_data_sync=1 --verify_checksum=1 --delete_obsolete_files_period_micros=314572800 --max_grandparent_overlap_factor=10 --max_background_compactions=4 --max_background_flushes=0 --level0_slowdown_writes_trigger=16 --level0_stop_writes_trigger=24 --statistics=0 --stats_per_interval=0 --stats_interval=1048576 --histogram=0 --use_plain_table=1 --open_files=-1 --mmap_read=1 --mmap_write=0 --memtablerep=prefix_hash --bloom_bits=10 --bloom_locality=1 --perf_level=0 --benchmarks=readwhilewriting,readwhilewriting,readwhilewriting --use_existing_db=1 --num=52428800 --threads=1 —writes_per_second=81920 Read Random: From 1.8363 ms/op, improve to 1.7587 ms/op. Read while writing: From 2.985 ms/op, improve to 2.924 ms/op. Test Plan: make all check Reviewers: ljin, haobo, yhchiang, sdong Reviewed By: sdong Subscribers: dhruba, igor Differential Revision: https://reviews.facebook.net/D19419
2014-07-10 07:14:39 +02:00
const FdWithKeyRange* f = &files[i];
const Slice file_limit = ExtractUserKey(f->largest_key);
if (user_cmp->Compare(file_limit, user_begin) >= 0) {
startIndex = i;
assert((count++, true));
} else {
break;
}
}
// check forward from 'mid+1' to higher indices
for (unsigned int i = midIndex+1;
i < level_files_brief_[level].num_files; i++) {
create compressed_levels_ in Version, allocate its space using arena. Make Version::Get, Version::FindFile faster Summary: Define CompressedFileMetaData that just contains fd, smallest_slice, largest_slice. Create compressed_levels_ in Version, the space is allocated using arena Thus increase the file meta data locality, speed up "Get" and "FindFile" benchmark with in-memory tmpfs, could have 4% improvement under "random read" and 2% improvement under "read while writing" benchmark command: ./db_bench --db=/mnt/db/rocksdb --num_levels=6 --key_size=20 --prefix_size=20 --keys_per_prefix=0 --value_size=100 --block_size=4096 --cache_size=17179869184 --cache_numshardbits=6 --compression_type=none --compression_ratio=1 --min_level_to_compress=-1 --disable_seek_compaction=1 --hard_rate_limit=2 --write_buffer_size=134217728 --max_write_buffer_number=2 --level0_file_num_compaction_trigger=8 --target_file_size_base=33554432 --max_bytes_for_level_base=1073741824 --disable_wal=0 --sync=0 --disable_data_sync=1 --verify_checksum=1 --delete_obsolete_files_period_micros=314572800 --max_grandparent_overlap_factor=10 --max_background_compactions=4 --max_background_flushes=0 --level0_slowdown_writes_trigger=16 --level0_stop_writes_trigger=24 --statistics=0 --stats_per_interval=0 --stats_interval=1048576 --histogram=0 --use_plain_table=1 --open_files=-1 --mmap_read=1 --mmap_write=0 --memtablerep=prefix_hash --bloom_bits=10 --bloom_locality=1 --perf_level=0 --benchmarks=readwhilewriting,readwhilewriting,readwhilewriting --use_existing_db=1 --num=52428800 --threads=1 —writes_per_second=81920 Read Random: From 1.8363 ms/op, improve to 1.7587 ms/op. Read while writing: From 2.985 ms/op, improve to 2.924 ms/op. Test Plan: make all check Reviewers: ljin, haobo, yhchiang, sdong Reviewed By: sdong Subscribers: dhruba, igor Differential Revision: https://reviews.facebook.net/D19419
2014-07-10 07:14:39 +02:00
const FdWithKeyRange* f = &files[i];
const Slice file_start = ExtractUserKey(f->smallest_key);
if (user_cmp->Compare(file_start, user_end) <= 0) {
assert((count++, true));
endIndex = i;
} else {
break;
}
}
assert(count == endIndex - startIndex + 1);
// insert overlapping files into vector
for (int i = startIndex; i <= endIndex; i++) {
FileMetaData* f = files_[level][i];
inputs->push_back(f);
}
}
[RocksDB] [MergeOperator] The new Merge Interface! Uses merge sequences. Summary: Here are the major changes to the Merge Interface. It has been expanded to handle cases where the MergeOperator is not associative. It does so by stacking up merge operations while scanning through the key history (i.e.: during Get() or Compaction), until a valid Put/Delete/end-of-history is encountered; it then applies all of the merge operations in the correct sequence starting with the base/sentinel value. I have also introduced an "AssociativeMerge" function which allows the user to take advantage of associative merge operations (such as in the case of counters). The implementation will always attempt to merge the operations/operands themselves together when they are encountered, and will resort to the "stacking" method if and only if the "associative-merge" fails. This implementation is conjectured to allow MergeOperator to handle the general case, while still providing the user with the ability to take advantage of certain efficiencies in their own merge-operator / data-structure. NOTE: This is a preliminary diff. This must still go through a lot of review, revision, and testing. Feedback welcome! Test Plan: -This is a preliminary diff. I have only just begun testing/debugging it. -I will be testing this with the existing MergeOperator use-cases and unit-tests (counters, string-append, and redis-lists) -I will be "desk-checking" and walking through the code with the help gdb. -I will find a way of stress-testing the new interface / implementation using db_bench, db_test, merge_test, and/or db_stress. -I will ensure that my tests cover all cases: Get-Memtable, Get-Immutable-Memtable, Get-from-Disk, Iterator-Range-Scan, Flush-Memtable-to-L0, Compaction-L0-L1, Compaction-Ln-L(n+1), Put/Delete found, Put/Delete not-found, end-of-history, end-of-file, etc. -A lot of feedback from the reviewers. Reviewers: haobo, dhruba, zshao, emayanke Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D11499
2013-08-06 05:14:32 +02:00
// Returns true iff the first or last file in inputs contains
// an overlapping user key to the file "just outside" of it (i.e.
// just after the last file, or just before the first file)
// REQUIRES: "*inputs" is a sorted list of non-overlapping files
bool VersionStorageInfo::HasOverlappingUserKey(
const std::vector<FileMetaData*>* inputs, int level) {
[RocksDB] [MergeOperator] The new Merge Interface! Uses merge sequences. Summary: Here are the major changes to the Merge Interface. It has been expanded to handle cases where the MergeOperator is not associative. It does so by stacking up merge operations while scanning through the key history (i.e.: during Get() or Compaction), until a valid Put/Delete/end-of-history is encountered; it then applies all of the merge operations in the correct sequence starting with the base/sentinel value. I have also introduced an "AssociativeMerge" function which allows the user to take advantage of associative merge operations (such as in the case of counters). The implementation will always attempt to merge the operations/operands themselves together when they are encountered, and will resort to the "stacking" method if and only if the "associative-merge" fails. This implementation is conjectured to allow MergeOperator to handle the general case, while still providing the user with the ability to take advantage of certain efficiencies in their own merge-operator / data-structure. NOTE: This is a preliminary diff. This must still go through a lot of review, revision, and testing. Feedback welcome! Test Plan: -This is a preliminary diff. I have only just begun testing/debugging it. -I will be testing this with the existing MergeOperator use-cases and unit-tests (counters, string-append, and redis-lists) -I will be "desk-checking" and walking through the code with the help gdb. -I will find a way of stress-testing the new interface / implementation using db_bench, db_test, merge_test, and/or db_stress. -I will ensure that my tests cover all cases: Get-Memtable, Get-Immutable-Memtable, Get-from-Disk, Iterator-Range-Scan, Flush-Memtable-to-L0, Compaction-L0-L1, Compaction-Ln-L(n+1), Put/Delete found, Put/Delete not-found, end-of-history, end-of-file, etc. -A lot of feedback from the reviewers. Reviewers: haobo, dhruba, zshao, emayanke Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D11499
2013-08-06 05:14:32 +02:00
// If inputs empty, there is no overlap.
// If level == 0, it is assumed that all needed files were already included.
if (inputs->empty() || level == 0){
return false;
}
const Comparator* user_cmp = user_comparator_;
const rocksdb::LevelFilesBrief& file_level = level_files_brief_[level];
const FdWithKeyRange* files = level_files_brief_[level].files;
create compressed_levels_ in Version, allocate its space using arena. Make Version::Get, Version::FindFile faster Summary: Define CompressedFileMetaData that just contains fd, smallest_slice, largest_slice. Create compressed_levels_ in Version, the space is allocated using arena Thus increase the file meta data locality, speed up "Get" and "FindFile" benchmark with in-memory tmpfs, could have 4% improvement under "random read" and 2% improvement under "read while writing" benchmark command: ./db_bench --db=/mnt/db/rocksdb --num_levels=6 --key_size=20 --prefix_size=20 --keys_per_prefix=0 --value_size=100 --block_size=4096 --cache_size=17179869184 --cache_numshardbits=6 --compression_type=none --compression_ratio=1 --min_level_to_compress=-1 --disable_seek_compaction=1 --hard_rate_limit=2 --write_buffer_size=134217728 --max_write_buffer_number=2 --level0_file_num_compaction_trigger=8 --target_file_size_base=33554432 --max_bytes_for_level_base=1073741824 --disable_wal=0 --sync=0 --disable_data_sync=1 --verify_checksum=1 --delete_obsolete_files_period_micros=314572800 --max_grandparent_overlap_factor=10 --max_background_compactions=4 --max_background_flushes=0 --level0_slowdown_writes_trigger=16 --level0_stop_writes_trigger=24 --statistics=0 --stats_per_interval=0 --stats_interval=1048576 --histogram=0 --use_plain_table=1 --open_files=-1 --mmap_read=1 --mmap_write=0 --memtablerep=prefix_hash --bloom_bits=10 --bloom_locality=1 --perf_level=0 --benchmarks=readwhilewriting,readwhilewriting,readwhilewriting --use_existing_db=1 --num=52428800 --threads=1 —writes_per_second=81920 Read Random: From 1.8363 ms/op, improve to 1.7587 ms/op. Read while writing: From 2.985 ms/op, improve to 2.924 ms/op. Test Plan: make all check Reviewers: ljin, haobo, yhchiang, sdong Reviewed By: sdong Subscribers: dhruba, igor Differential Revision: https://reviews.facebook.net/D19419
2014-07-10 07:14:39 +02:00
const size_t kNumFiles = file_level.num_files;
[RocksDB] [MergeOperator] The new Merge Interface! Uses merge sequences. Summary: Here are the major changes to the Merge Interface. It has been expanded to handle cases where the MergeOperator is not associative. It does so by stacking up merge operations while scanning through the key history (i.e.: during Get() or Compaction), until a valid Put/Delete/end-of-history is encountered; it then applies all of the merge operations in the correct sequence starting with the base/sentinel value. I have also introduced an "AssociativeMerge" function which allows the user to take advantage of associative merge operations (such as in the case of counters). The implementation will always attempt to merge the operations/operands themselves together when they are encountered, and will resort to the "stacking" method if and only if the "associative-merge" fails. This implementation is conjectured to allow MergeOperator to handle the general case, while still providing the user with the ability to take advantage of certain efficiencies in their own merge-operator / data-structure. NOTE: This is a preliminary diff. This must still go through a lot of review, revision, and testing. Feedback welcome! Test Plan: -This is a preliminary diff. I have only just begun testing/debugging it. -I will be testing this with the existing MergeOperator use-cases and unit-tests (counters, string-append, and redis-lists) -I will be "desk-checking" and walking through the code with the help gdb. -I will find a way of stress-testing the new interface / implementation using db_bench, db_test, merge_test, and/or db_stress. -I will ensure that my tests cover all cases: Get-Memtable, Get-Immutable-Memtable, Get-from-Disk, Iterator-Range-Scan, Flush-Memtable-to-L0, Compaction-L0-L1, Compaction-Ln-L(n+1), Put/Delete found, Put/Delete not-found, end-of-history, end-of-file, etc. -A lot of feedback from the reviewers. Reviewers: haobo, dhruba, zshao, emayanke Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D11499
2013-08-06 05:14:32 +02:00
// Check the last file in inputs against the file after it
size_t last_file = FindFile(*internal_comparator_, file_level,
[RocksDB] [MergeOperator] The new Merge Interface! Uses merge sequences. Summary: Here are the major changes to the Merge Interface. It has been expanded to handle cases where the MergeOperator is not associative. It does so by stacking up merge operations while scanning through the key history (i.e.: during Get() or Compaction), until a valid Put/Delete/end-of-history is encountered; it then applies all of the merge operations in the correct sequence starting with the base/sentinel value. I have also introduced an "AssociativeMerge" function which allows the user to take advantage of associative merge operations (such as in the case of counters). The implementation will always attempt to merge the operations/operands themselves together when they are encountered, and will resort to the "stacking" method if and only if the "associative-merge" fails. This implementation is conjectured to allow MergeOperator to handle the general case, while still providing the user with the ability to take advantage of certain efficiencies in their own merge-operator / data-structure. NOTE: This is a preliminary diff. This must still go through a lot of review, revision, and testing. Feedback welcome! Test Plan: -This is a preliminary diff. I have only just begun testing/debugging it. -I will be testing this with the existing MergeOperator use-cases and unit-tests (counters, string-append, and redis-lists) -I will be "desk-checking" and walking through the code with the help gdb. -I will find a way of stress-testing the new interface / implementation using db_bench, db_test, merge_test, and/or db_stress. -I will ensure that my tests cover all cases: Get-Memtable, Get-Immutable-Memtable, Get-from-Disk, Iterator-Range-Scan, Flush-Memtable-to-L0, Compaction-L0-L1, Compaction-Ln-L(n+1), Put/Delete found, Put/Delete not-found, end-of-history, end-of-file, etc. -A lot of feedback from the reviewers. Reviewers: haobo, dhruba, zshao, emayanke Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D11499
2013-08-06 05:14:32 +02:00
inputs->back()->largest.Encode());
assert(last_file < kNumFiles); // File should exist!
[RocksDB] [MergeOperator] The new Merge Interface! Uses merge sequences. Summary: Here are the major changes to the Merge Interface. It has been expanded to handle cases where the MergeOperator is not associative. It does so by stacking up merge operations while scanning through the key history (i.e.: during Get() or Compaction), until a valid Put/Delete/end-of-history is encountered; it then applies all of the merge operations in the correct sequence starting with the base/sentinel value. I have also introduced an "AssociativeMerge" function which allows the user to take advantage of associative merge operations (such as in the case of counters). The implementation will always attempt to merge the operations/operands themselves together when they are encountered, and will resort to the "stacking" method if and only if the "associative-merge" fails. This implementation is conjectured to allow MergeOperator to handle the general case, while still providing the user with the ability to take advantage of certain efficiencies in their own merge-operator / data-structure. NOTE: This is a preliminary diff. This must still go through a lot of review, revision, and testing. Feedback welcome! Test Plan: -This is a preliminary diff. I have only just begun testing/debugging it. -I will be testing this with the existing MergeOperator use-cases and unit-tests (counters, string-append, and redis-lists) -I will be "desk-checking" and walking through the code with the help gdb. -I will find a way of stress-testing the new interface / implementation using db_bench, db_test, merge_test, and/or db_stress. -I will ensure that my tests cover all cases: Get-Memtable, Get-Immutable-Memtable, Get-from-Disk, Iterator-Range-Scan, Flush-Memtable-to-L0, Compaction-L0-L1, Compaction-Ln-L(n+1), Put/Delete found, Put/Delete not-found, end-of-history, end-of-file, etc. -A lot of feedback from the reviewers. Reviewers: haobo, dhruba, zshao, emayanke Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D11499
2013-08-06 05:14:32 +02:00
if (last_file < kNumFiles-1) { // If not the last file
create compressed_levels_ in Version, allocate its space using arena. Make Version::Get, Version::FindFile faster Summary: Define CompressedFileMetaData that just contains fd, smallest_slice, largest_slice. Create compressed_levels_ in Version, the space is allocated using arena Thus increase the file meta data locality, speed up "Get" and "FindFile" benchmark with in-memory tmpfs, could have 4% improvement under "random read" and 2% improvement under "read while writing" benchmark command: ./db_bench --db=/mnt/db/rocksdb --num_levels=6 --key_size=20 --prefix_size=20 --keys_per_prefix=0 --value_size=100 --block_size=4096 --cache_size=17179869184 --cache_numshardbits=6 --compression_type=none --compression_ratio=1 --min_level_to_compress=-1 --disable_seek_compaction=1 --hard_rate_limit=2 --write_buffer_size=134217728 --max_write_buffer_number=2 --level0_file_num_compaction_trigger=8 --target_file_size_base=33554432 --max_bytes_for_level_base=1073741824 --disable_wal=0 --sync=0 --disable_data_sync=1 --verify_checksum=1 --delete_obsolete_files_period_micros=314572800 --max_grandparent_overlap_factor=10 --max_background_compactions=4 --max_background_flushes=0 --level0_slowdown_writes_trigger=16 --level0_stop_writes_trigger=24 --statistics=0 --stats_per_interval=0 --stats_interval=1048576 --histogram=0 --use_plain_table=1 --open_files=-1 --mmap_read=1 --mmap_write=0 --memtablerep=prefix_hash --bloom_bits=10 --bloom_locality=1 --perf_level=0 --benchmarks=readwhilewriting,readwhilewriting,readwhilewriting --use_existing_db=1 --num=52428800 --threads=1 —writes_per_second=81920 Read Random: From 1.8363 ms/op, improve to 1.7587 ms/op. Read while writing: From 2.985 ms/op, improve to 2.924 ms/op. Test Plan: make all check Reviewers: ljin, haobo, yhchiang, sdong Reviewed By: sdong Subscribers: dhruba, igor Differential Revision: https://reviews.facebook.net/D19419
2014-07-10 07:14:39 +02:00
const Slice last_key_in_input = ExtractUserKey(
files[last_file].largest_key);
const Slice first_key_after = ExtractUserKey(
files[last_file+1].smallest_key);
[RocksDB] [MergeOperator] The new Merge Interface! Uses merge sequences. Summary: Here are the major changes to the Merge Interface. It has been expanded to handle cases where the MergeOperator is not associative. It does so by stacking up merge operations while scanning through the key history (i.e.: during Get() or Compaction), until a valid Put/Delete/end-of-history is encountered; it then applies all of the merge operations in the correct sequence starting with the base/sentinel value. I have also introduced an "AssociativeMerge" function which allows the user to take advantage of associative merge operations (such as in the case of counters). The implementation will always attempt to merge the operations/operands themselves together when they are encountered, and will resort to the "stacking" method if and only if the "associative-merge" fails. This implementation is conjectured to allow MergeOperator to handle the general case, while still providing the user with the ability to take advantage of certain efficiencies in their own merge-operator / data-structure. NOTE: This is a preliminary diff. This must still go through a lot of review, revision, and testing. Feedback welcome! Test Plan: -This is a preliminary diff. I have only just begun testing/debugging it. -I will be testing this with the existing MergeOperator use-cases and unit-tests (counters, string-append, and redis-lists) -I will be "desk-checking" and walking through the code with the help gdb. -I will find a way of stress-testing the new interface / implementation using db_bench, db_test, merge_test, and/or db_stress. -I will ensure that my tests cover all cases: Get-Memtable, Get-Immutable-Memtable, Get-from-Disk, Iterator-Range-Scan, Flush-Memtable-to-L0, Compaction-L0-L1, Compaction-Ln-L(n+1), Put/Delete found, Put/Delete not-found, end-of-history, end-of-file, etc. -A lot of feedback from the reviewers. Reviewers: haobo, dhruba, zshao, emayanke Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D11499
2013-08-06 05:14:32 +02:00
if (user_cmp->Compare(last_key_in_input, first_key_after) == 0) {
// The last user key in input overlaps with the next file's first key
return true;
}
}
// Check the first file in inputs against the file just before it
size_t first_file = FindFile(*internal_comparator_, file_level,
[RocksDB] [MergeOperator] The new Merge Interface! Uses merge sequences. Summary: Here are the major changes to the Merge Interface. It has been expanded to handle cases where the MergeOperator is not associative. It does so by stacking up merge operations while scanning through the key history (i.e.: during Get() or Compaction), until a valid Put/Delete/end-of-history is encountered; it then applies all of the merge operations in the correct sequence starting with the base/sentinel value. I have also introduced an "AssociativeMerge" function which allows the user to take advantage of associative merge operations (such as in the case of counters). The implementation will always attempt to merge the operations/operands themselves together when they are encountered, and will resort to the "stacking" method if and only if the "associative-merge" fails. This implementation is conjectured to allow MergeOperator to handle the general case, while still providing the user with the ability to take advantage of certain efficiencies in their own merge-operator / data-structure. NOTE: This is a preliminary diff. This must still go through a lot of review, revision, and testing. Feedback welcome! Test Plan: -This is a preliminary diff. I have only just begun testing/debugging it. -I will be testing this with the existing MergeOperator use-cases and unit-tests (counters, string-append, and redis-lists) -I will be "desk-checking" and walking through the code with the help gdb. -I will find a way of stress-testing the new interface / implementation using db_bench, db_test, merge_test, and/or db_stress. -I will ensure that my tests cover all cases: Get-Memtable, Get-Immutable-Memtable, Get-from-Disk, Iterator-Range-Scan, Flush-Memtable-to-L0, Compaction-L0-L1, Compaction-Ln-L(n+1), Put/Delete found, Put/Delete not-found, end-of-history, end-of-file, etc. -A lot of feedback from the reviewers. Reviewers: haobo, dhruba, zshao, emayanke Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D11499
2013-08-06 05:14:32 +02:00
inputs->front()->smallest.Encode());
assert(first_file <= last_file); // File should exist!
[RocksDB] [MergeOperator] The new Merge Interface! Uses merge sequences. Summary: Here are the major changes to the Merge Interface. It has been expanded to handle cases where the MergeOperator is not associative. It does so by stacking up merge operations while scanning through the key history (i.e.: during Get() or Compaction), until a valid Put/Delete/end-of-history is encountered; it then applies all of the merge operations in the correct sequence starting with the base/sentinel value. I have also introduced an "AssociativeMerge" function which allows the user to take advantage of associative merge operations (such as in the case of counters). The implementation will always attempt to merge the operations/operands themselves together when they are encountered, and will resort to the "stacking" method if and only if the "associative-merge" fails. This implementation is conjectured to allow MergeOperator to handle the general case, while still providing the user with the ability to take advantage of certain efficiencies in their own merge-operator / data-structure. NOTE: This is a preliminary diff. This must still go through a lot of review, revision, and testing. Feedback welcome! Test Plan: -This is a preliminary diff. I have only just begun testing/debugging it. -I will be testing this with the existing MergeOperator use-cases and unit-tests (counters, string-append, and redis-lists) -I will be "desk-checking" and walking through the code with the help gdb. -I will find a way of stress-testing the new interface / implementation using db_bench, db_test, merge_test, and/or db_stress. -I will ensure that my tests cover all cases: Get-Memtable, Get-Immutable-Memtable, Get-from-Disk, Iterator-Range-Scan, Flush-Memtable-to-L0, Compaction-L0-L1, Compaction-Ln-L(n+1), Put/Delete found, Put/Delete not-found, end-of-history, end-of-file, etc. -A lot of feedback from the reviewers. Reviewers: haobo, dhruba, zshao, emayanke Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D11499
2013-08-06 05:14:32 +02:00
if (first_file > 0) { // If not first file
create compressed_levels_ in Version, allocate its space using arena. Make Version::Get, Version::FindFile faster Summary: Define CompressedFileMetaData that just contains fd, smallest_slice, largest_slice. Create compressed_levels_ in Version, the space is allocated using arena Thus increase the file meta data locality, speed up "Get" and "FindFile" benchmark with in-memory tmpfs, could have 4% improvement under "random read" and 2% improvement under "read while writing" benchmark command: ./db_bench --db=/mnt/db/rocksdb --num_levels=6 --key_size=20 --prefix_size=20 --keys_per_prefix=0 --value_size=100 --block_size=4096 --cache_size=17179869184 --cache_numshardbits=6 --compression_type=none --compression_ratio=1 --min_level_to_compress=-1 --disable_seek_compaction=1 --hard_rate_limit=2 --write_buffer_size=134217728 --max_write_buffer_number=2 --level0_file_num_compaction_trigger=8 --target_file_size_base=33554432 --max_bytes_for_level_base=1073741824 --disable_wal=0 --sync=0 --disable_data_sync=1 --verify_checksum=1 --delete_obsolete_files_period_micros=314572800 --max_grandparent_overlap_factor=10 --max_background_compactions=4 --max_background_flushes=0 --level0_slowdown_writes_trigger=16 --level0_stop_writes_trigger=24 --statistics=0 --stats_per_interval=0 --stats_interval=1048576 --histogram=0 --use_plain_table=1 --open_files=-1 --mmap_read=1 --mmap_write=0 --memtablerep=prefix_hash --bloom_bits=10 --bloom_locality=1 --perf_level=0 --benchmarks=readwhilewriting,readwhilewriting,readwhilewriting --use_existing_db=1 --num=52428800 --threads=1 —writes_per_second=81920 Read Random: From 1.8363 ms/op, improve to 1.7587 ms/op. Read while writing: From 2.985 ms/op, improve to 2.924 ms/op. Test Plan: make all check Reviewers: ljin, haobo, yhchiang, sdong Reviewed By: sdong Subscribers: dhruba, igor Differential Revision: https://reviews.facebook.net/D19419
2014-07-10 07:14:39 +02:00
const Slice& first_key_in_input = ExtractUserKey(
files[first_file].smallest_key);
const Slice& last_key_before = ExtractUserKey(
files[first_file-1].largest_key);
[RocksDB] [MergeOperator] The new Merge Interface! Uses merge sequences. Summary: Here are the major changes to the Merge Interface. It has been expanded to handle cases where the MergeOperator is not associative. It does so by stacking up merge operations while scanning through the key history (i.e.: during Get() or Compaction), until a valid Put/Delete/end-of-history is encountered; it then applies all of the merge operations in the correct sequence starting with the base/sentinel value. I have also introduced an "AssociativeMerge" function which allows the user to take advantage of associative merge operations (such as in the case of counters). The implementation will always attempt to merge the operations/operands themselves together when they are encountered, and will resort to the "stacking" method if and only if the "associative-merge" fails. This implementation is conjectured to allow MergeOperator to handle the general case, while still providing the user with the ability to take advantage of certain efficiencies in their own merge-operator / data-structure. NOTE: This is a preliminary diff. This must still go through a lot of review, revision, and testing. Feedback welcome! Test Plan: -This is a preliminary diff. I have only just begun testing/debugging it. -I will be testing this with the existing MergeOperator use-cases and unit-tests (counters, string-append, and redis-lists) -I will be "desk-checking" and walking through the code with the help gdb. -I will find a way of stress-testing the new interface / implementation using db_bench, db_test, merge_test, and/or db_stress. -I will ensure that my tests cover all cases: Get-Memtable, Get-Immutable-Memtable, Get-from-Disk, Iterator-Range-Scan, Flush-Memtable-to-L0, Compaction-L0-L1, Compaction-Ln-L(n+1), Put/Delete found, Put/Delete not-found, end-of-history, end-of-file, etc. -A lot of feedback from the reviewers. Reviewers: haobo, dhruba, zshao, emayanke Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D11499
2013-08-06 05:14:32 +02:00
if (user_cmp->Compare(first_key_in_input, last_key_before) == 0) {
// The first user key in input overlaps with the previous file's last key
return true;
}
}
return false;
}
uint64_t VersionStorageInfo::NumLevelBytes(int level) const {
assert(level >= 0);
assert(level < num_levels());
return TotalFileSize(files_[level]);
}
const char* VersionStorageInfo::LevelSummary(
LevelSummaryStorage* scratch) const {
int len = snprintf(scratch->buffer, sizeof(scratch->buffer), "files[");
for (int i = 0; i < num_levels(); i++) {
int sz = sizeof(scratch->buffer) - len;
int ret = snprintf(scratch->buffer + len, sz, "%d ", int(files_[i].size()));
if (ret < 0 || ret >= sz) break;
len += ret;
}
if (len > 0) {
// overwrite the last space
--len;
}
snprintf(scratch->buffer + len, sizeof(scratch->buffer) - len, "]");
return scratch->buffer;
}
const char* VersionStorageInfo::LevelFileSummary(FileSummaryStorage* scratch,
int level) const {
int len = snprintf(scratch->buffer, sizeof(scratch->buffer), "files_size[");
for (const auto& f : files_[level]) {
int sz = sizeof(scratch->buffer) - len;
char sztxt[16];
AppendHumanBytes(f->fd.GetFileSize(), sztxt, sizeof(sztxt));
int ret = snprintf(scratch->buffer + len, sz,
"#%" PRIu64 "(seq=%" PRIu64 ",sz=%s,%d) ",
f->fd.GetNumber(), f->smallest_seqno, sztxt,
static_cast<int>(f->being_compacted));
if (ret < 0 || ret >= sz)
break;
len += ret;
}
// overwrite the last space (only if files_[level].size() is non-zero)
if (files_[level].size() && len > 0) {
--len;
}
snprintf(scratch->buffer + len, sizeof(scratch->buffer) - len, "]");
return scratch->buffer;
}
int64_t VersionStorageInfo::MaxNextLevelOverlappingBytes() {
uint64_t result = 0;
std::vector<FileMetaData*> overlaps;
for (int level = 1; level < num_levels() - 1; level++) {
for (const auto& f : files_[level]) {
GetOverlappingInputs(level + 1, &f->smallest, &f->largest, &overlaps);
const uint64_t sum = TotalFileSize(overlaps);
if (sum > result) {
result = sum;
}
}
}
return result;
}
void Version::AddLiveFiles(std::vector<FileDescriptor>* live) {
for (int level = 0; level < storage_info_.num_levels(); level++) {
const std::vector<FileMetaData*>& files = storage_info_.files_[level];
for (const auto& file : files) {
live->push_back(file->fd);
}
}
}
std::string Version::DebugString(bool hex) const {
std::string r;
for (int level = 0; level < storage_info_.num_levels_; level++) {
// E.g.,
// --- level 1 ---
// 17:123['a' .. 'd']
// 20:43['e' .. 'g']
r.append("--- level ");
AppendNumberTo(&r, level);
r.append(" --- version# ");
AppendNumberTo(&r, version_number_);
r.append(" ---\n");
const std::vector<FileMetaData*>& files = storage_info_.files_[level];
for (size_t i = 0; i < files.size(); i++) {
r.push_back(' ');
AppendNumberTo(&r, files[i]->fd.GetNumber());
r.push_back(':');
AppendNumberTo(&r, files[i]->fd.GetFileSize());
r.append("[");
r.append(files[i]->smallest.DebugString(hex));
r.append(" .. ");
r.append(files[i]->largest.DebugString(hex));
r.append("]\n");
}
}
return r;
}
// this is used to batch writes to the manifest file
struct VersionSet::ManifestWriter {
Status status;
bool done;
port::CondVar cv;
ColumnFamilyData* cfd;
VersionEdit* edit;
explicit ManifestWriter(port::Mutex* mu, ColumnFamilyData* _cfd,
VersionEdit* e)
: done(false), cv(mu), cfd(_cfd), edit(e) {}
};
VersionSet::VersionSet(const std::string& dbname, const DBOptions* db_options,
const EnvOptions& storage_options, Cache* table_cache,
WriteBuffer* write_buffer,
WriteController* write_controller)
: column_family_set_(new ColumnFamilySet(
dbname, db_options, storage_options, table_cache,
write_buffer, write_controller)),
env_(db_options->env),
dbname_(dbname),
db_options_(db_options),
next_file_number_(2),
manifest_file_number_(0), // Filled by Recover()
pending_manifest_file_number_(0),
last_sequence_(0),
prev_log_number_(0),
current_version_number_(0),
manifest_file_size_(0),
env_options_(storage_options),
env_options_compactions_(env_options_) {}
VersionSet::~VersionSet() {
// we need to delete column_family_set_ because its destructor depends on
// VersionSet
column_family_set_.reset();
for (auto file : obsolete_files_) {
delete file;
}
obsolete_files_.clear();
}
void VersionSet::AppendVersion(ColumnFamilyData* column_family_data,
Version* v) {
// Mark v finalized
v->storage_info_.SetFinalized();
// Make "v" current
assert(v->refs_ == 0);
Version* current = column_family_data->current();
assert(v != current);
if (current != nullptr) {
assert(current->refs_ > 0);
current->Unref();
}
column_family_data->SetCurrent(v);
v->Ref();
// Append to linked list
v->prev_ = column_family_data->dummy_versions()->prev_;
v->next_ = column_family_data->dummy_versions();
v->prev_->next_ = v;
v->next_->prev_ = v;
}
Status VersionSet::LogAndApply(ColumnFamilyData* column_family_data,
const MutableCFOptions& mutable_cf_options,
VersionEdit* edit, port::Mutex* mu,
Directory* db_directory, bool new_descriptor_log,
const ColumnFamilyOptions* new_cf_options) {
mu->AssertHeld();
2014-03-13 02:09:03 +01:00
// column_family_data can be nullptr only if this is column_family_add.
// in that case, we also need to specify ColumnFamilyOptions
if (column_family_data == nullptr) {
assert(edit->is_column_family_add_);
assert(new_cf_options != nullptr);
}
// queue our request
ManifestWriter w(mu, column_family_data, edit);
manifest_writers_.push_back(&w);
while (!w.done && &w != manifest_writers_.front()) {
w.cv.Wait();
}
if (w.done) {
return w.status;
}
if (column_family_data != nullptr && column_family_data->IsDropped()) {
// if column family is dropped by the time we get here, no need to write
// anything to the manifest
manifest_writers_.pop_front();
// Notify new head of write queue
if (!manifest_writers_.empty()) {
manifest_writers_.front()->cv.Signal();
}
return Status::OK();
}
std::vector<VersionEdit*> batch_edits;
Version* v = nullptr;
std::unique_ptr<BaseReferencedVersionBuilder> builder_guard(nullptr);
// process all requests in the queue
ManifestWriter* last_writer = &w;
assert(!manifest_writers_.empty());
assert(manifest_writers_.front() == &w);
if (edit->IsColumnFamilyManipulation()) {
// no group commits for column family add or drop
2014-03-13 02:09:03 +01:00
LogAndApplyCFHelper(edit);
batch_edits.push_back(edit);
} else {
v = new Version(column_family_data, this, current_version_number_++);
builder_guard.reset(new BaseReferencedVersionBuilder(column_family_data));
auto* builder = builder_guard->version_builder();
for (const auto& writer : manifest_writers_) {
if (writer->edit->IsColumnFamilyManipulation() ||
writer->cfd->GetID() != column_family_data->GetID()) {
// no group commits for column family add or drop
// also, group commits across column families are not supported
break;
}
last_writer = writer;
LogAndApplyHelper(column_family_data, builder, v, last_writer->edit, mu);
batch_edits.push_back(last_writer->edit);
}
builder->SaveTo(v->storage_info());
}
// Initialize new descriptor log file if necessary by creating
// a temporary file that contains a snapshot of the current version.
uint64_t new_manifest_file_size = 0;
Status s;
assert(pending_manifest_file_number_ == 0);
if (!descriptor_log_ ||
manifest_file_size_ > db_options_->max_manifest_file_size) {
pending_manifest_file_number_ = NewFileNumber();
batch_edits.back()->SetNextFile(next_file_number_.load());
new_descriptor_log = true;
} else {
pending_manifest_file_number_ = manifest_file_number_;
}
if (new_descriptor_log) {
// if we're writing out new snapshot make sure to persist max column family
2014-03-13 02:09:03 +01:00
if (column_family_set_->GetMaxColumnFamily() > 0) {
edit->SetMaxColumnFamily(column_family_set_->GetMaxColumnFamily());
}
}
// Unlock during expensive operations. New writes cannot get here
// because &w is ensuring that all new writes get queued.
{
std::vector<uint64_t> size_being_compacted;
if (!edit->IsColumnFamilyManipulation()) {
size_being_compacted.resize(v->storage_info()->num_levels() - 1);
// calculate the amount of data being compacted at every level
column_family_data->compaction_picker()->SizeBeingCompacted(
size_being_compacted);
}
Prevent segfault because SizeUnderCompaction was called without any locks. Summary: SizeBeingCompacted was called without any lock protection. This causes crashes, especially when running db_bench with value_size=128K. The fix is to compute SizeUnderCompaction while holding the mutex and passing in these values into the call to Finalize. (gdb) where #4 leveldb::VersionSet::SizeBeingCompacted (this=this@entry=0x7f0b490931c0, level=level@entry=4) at db/version_set.cc:1827 #5 0x000000000043a3c8 in leveldb::VersionSet::Finalize (this=this@entry=0x7f0b490931c0, v=v@entry=0x7f0b3b86b480) at db/version_set.cc:1420 #6 0x00000000004418d1 in leveldb::VersionSet::LogAndApply (this=0x7f0b490931c0, edit=0x7f0b3dc8c200, mu=0x7f0b490835b0, new_descriptor_log=<optimized out>) at db/version_set.cc:1016 #7 0x00000000004222b2 in leveldb::DBImpl::InstallCompactionResults (this=this@entry=0x7f0b49083400, compact=compact@entry=0x7f0b2b8330f0) at db/db_impl.cc:1473 #8 0x0000000000426027 in leveldb::DBImpl::DoCompactionWork (this=this@entry=0x7f0b49083400, compact=compact@entry=0x7f0b2b8330f0) at db/db_impl.cc:1757 #9 0x0000000000426690 in leveldb::DBImpl::BackgroundCompaction (this=this@entry=0x7f0b49083400, madeProgress=madeProgress@entry=0x7f0b41bf2d1e, deletion_state=...) at db/db_impl.cc:1268 #10 0x0000000000428f42 in leveldb::DBImpl::BackgroundCall (this=0x7f0b49083400) at db/db_impl.cc:1170 #11 0x000000000045348e in BGThread (this=0x7f0b49023100) at util/env_posix.cc:941 #12 leveldb::(anonymous namespace)::PosixEnv::BGThreadWrapper (arg=0x7f0b49023100) at util/env_posix.cc:874 #13 0x00007f0b4a7cf10d in start_thread (arg=0x7f0b41bf3700) at pthread_create.c:301 #14 0x00007f0b49b4b11d in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:115 Test Plan: make check I am running db_bench with a value size of 128K to see if the segfault is fixed. Reviewers: MarkCallaghan, sheki, emayanke Reviewed By: sheki CC: leveldb Differential Revision: https://reviews.facebook.net/D9279
2013-03-11 17:47:48 +01:00
mu->Unlock();
if (!edit->IsColumnFamilyManipulation() &&
db_options_->max_open_files == -1) {
// unlimited table cache. Pre-load table handle now.
// Need to do it out of the mutex.
builder_guard->version_builder()->LoadTableHandlers();
}
// This is fine because everything inside of this block is serialized --
// only one thread can be here at the same time
2014-03-13 02:09:03 +01:00
if (new_descriptor_log) {
// create manifest file
Log(InfoLogLevel::INFO_LEVEL, db_options_->info_log,
"Creating manifest %" PRIu64 "\n", pending_manifest_file_number_);
unique_ptr<WritableFile> descriptor_file;
s = env_->NewWritableFile(
DescriptorFileName(dbname_, pending_manifest_file_number_),
&descriptor_file, env_->OptimizeForManifestWrite(env_options_));
if (s.ok()) {
descriptor_file->SetPreallocationBlockSize(
db_options_->manifest_preallocation_size);
descriptor_log_.reset(new log::Writer(std::move(descriptor_file)));
s = WriteSnapshot(descriptor_log_.get());
}
}
if (!edit->IsColumnFamilyManipulation()) {
// This is cpu-heavy operations, which should be called outside mutex.
v->PrepareApply(mutable_cf_options, size_being_compacted);
}
// Write new record to MANIFEST log
if (s.ok()) {
2014-02-28 21:22:45 +01:00
for (auto& e : batch_edits) {
std::string record;
if (!e->EncodeTo(&record)) {
s = Status::Corruption(
"Unable to Encode VersionEdit:" + e->DebugString(true));
break;
}
s = descriptor_log_->AddRecord(record);
if (!s.ok()) {
break;
}
}
if (s.ok()) {
s = SyncManifest(env_, db_options_, descriptor_log_->file());
}
if (!s.ok()) {
Log(InfoLogLevel::ERROR_LEVEL, db_options_->info_log,
"MANIFEST write: %s\n", s.ToString().c_str());
2014-02-28 21:22:45 +01:00
bool all_records_in = true;
for (auto& e : batch_edits) {
std::string record;
if (!e->EncodeTo(&record)) {
s = Status::Corruption(
"Unable to Encode VersionEdit:" + e->DebugString(true));
all_records_in = false;
break;
}
if (!ManifestContains(pending_manifest_file_number_, record)) {
2014-02-28 21:22:45 +01:00
all_records_in = false;
break;
}
}
if (all_records_in) {
Log(InfoLogLevel::WARN_LEVEL, db_options_->info_log,
"MANIFEST contains log record despite error; advancing to new "
"version to prevent mismatch between in-memory and logged state"
" If paranoid is set, then the db is now in readonly mode.");
s = Status::OK();
}
}
}
// If we just created a new descriptor file, install it by writing a
// new CURRENT file that points to it.
if (s.ok() && new_descriptor_log) {
s = SetCurrentFile(env_, dbname_, pending_manifest_file_number_,
db_options_->disableDataSync ? nullptr : db_directory);
if (s.ok() && pending_manifest_file_number_ > manifest_file_number_) {
// delete old manifest file
Log(InfoLogLevel::INFO_LEVEL, db_options_->info_log,
"Deleting manifest %" PRIu64 " current manifest %" PRIu64 "\n",
manifest_file_number_, pending_manifest_file_number_);
// we don't care about an error here, PurgeObsoleteFiles will take care
// of it later
env_->DeleteFile(DescriptorFileName(dbname_, manifest_file_number_));
}
}
if (s.ok()) {
// find offset in manifest file where this version is stored.
new_manifest_file_size = descriptor_log_->file()->GetFileSize();
}
LogFlush(db_options_->info_log);
mu->Lock();
}
// Install the new version
if (s.ok()) {
if (edit->is_column_family_add_) {
// no group commit on column family add
assert(batch_edits.size() == 1);
assert(new_cf_options != nullptr);
CreateColumnFamily(*new_cf_options, edit);
} else if (edit->is_column_family_drop_) {
assert(batch_edits.size() == 1);
column_family_data->SetDropped();
if (column_family_data->Unref()) {
delete column_family_data;
}
} else {
2014-03-14 21:11:41 +01:00
uint64_t max_log_number_in_batch = 0;
for (auto& e : batch_edits) {
if (e->has_log_number_) {
max_log_number_in_batch =
std::max(max_log_number_in_batch, e->log_number_);
}
}
if (max_log_number_in_batch != 0) {
assert(column_family_data->GetLogNumber() <= max_log_number_in_batch);
2014-03-14 21:11:41 +01:00
column_family_data->SetLogNumber(max_log_number_in_batch);
}
AppendVersion(column_family_data, v);
}
manifest_file_number_ = pending_manifest_file_number_;
manifest_file_size_ = new_manifest_file_size;
prev_log_number_ = edit->prev_log_number_;
} else {
Log(InfoLogLevel::ERROR_LEVEL, db_options_->info_log,
"Error in committing version %lu to [%s]",
(unsigned long)v->GetVersionNumber(),
column_family_data->GetName().c_str());
delete v;
if (new_descriptor_log) {
Log(InfoLogLevel::INFO_LEVEL, db_options_->info_log,
"Deleting manifest %" PRIu64 " current manifest %" PRIu64 "\n",
manifest_file_number_, pending_manifest_file_number_);
descriptor_log_.reset();
env_->DeleteFile(
DescriptorFileName(dbname_, pending_manifest_file_number_));
}
}
pending_manifest_file_number_ = 0;
// wake up all the waiting writers
while (true) {
ManifestWriter* ready = manifest_writers_.front();
manifest_writers_.pop_front();
if (ready != &w) {
ready->status = s;
ready->done = true;
ready->cv.Signal();
}
if (ready == last_writer) break;
}
// Notify new head of write queue
if (!manifest_writers_.empty()) {
manifest_writers_.front()->cv.Signal();
}
return s;
}
2014-03-13 02:09:03 +01:00
void VersionSet::LogAndApplyCFHelper(VersionEdit* edit) {
assert(edit->IsColumnFamilyManipulation());
edit->SetNextFile(next_file_number_.load());
2014-03-13 02:09:03 +01:00
edit->SetLastSequence(last_sequence_);
if (edit->is_column_family_drop_) {
// if we drop column family, we have to make sure to save max column family,
// so that we don't reuse existing ID
edit->SetMaxColumnFamily(column_family_set_->GetMaxColumnFamily());
}
}
void VersionSet::LogAndApplyHelper(ColumnFamilyData* cfd,
VersionBuilder* builder, Version* v,
VersionEdit* edit, port::Mutex* mu) {
mu->AssertHeld();
2014-03-13 02:09:03 +01:00
assert(!edit->IsColumnFamilyManipulation());
if (edit->has_log_number_) {
assert(edit->log_number_ >= cfd->GetLogNumber());
assert(edit->log_number_ < next_file_number_.load());
}
2014-03-13 02:09:03 +01:00
if (!edit->has_prev_log_number_) {
edit->SetPrevLogNumber(prev_log_number_);
}
edit->SetNextFile(next_file_number_.load());
2014-03-13 02:09:03 +01:00
edit->SetLastSequence(last_sequence_);
builder->Apply(edit);
}
Status VersionSet::Recover(
const std::vector<ColumnFamilyDescriptor>& column_families,
bool read_only) {
std::unordered_map<std::string, ColumnFamilyOptions> cf_name_to_options;
for (auto cf : column_families) {
cf_name_to_options.insert({cf.name, cf.options});
}
// keeps track of column families in manifest that were not found in
// column families parameters. if those column families are not dropped
// by subsequent manifest records, Recover() will return failure status
2014-04-09 19:38:05 +02:00
std::unordered_map<int, std::string> column_families_not_found;
// Read "CURRENT" file, which contains a pointer to the current manifest file
std::string manifest_filename;
Status s = ReadFileToString(
env_, CurrentFileName(dbname_), &manifest_filename
);
if (!s.ok()) {
return s;
}
if (manifest_filename.empty() ||
manifest_filename.back() != '\n') {
return Status::Corruption("CURRENT file does not end with newline");
}
// remove the trailing '\n'
manifest_filename.resize(manifest_filename.size() - 1);
2014-03-12 18:52:32 +01:00
FileType type;
bool parse_ok =
ParseFileName(manifest_filename, &manifest_file_number_, &type);
if (!parse_ok || type != kDescriptorFile) {
return Status::Corruption("CURRENT file corrupted");
}
Log(InfoLogLevel::INFO_LEVEL, db_options_->info_log,
"Recovering from manifest file: %s\n",
manifest_filename.c_str());
manifest_filename = dbname_ + "/" + manifest_filename;
unique_ptr<SequentialFile> manifest_file;
2014-03-12 18:52:32 +01:00
s = env_->NewSequentialFile(manifest_filename, &manifest_file,
env_options_);
if (!s.ok()) {
return s;
}
uint64_t current_manifest_file_size;
s = env_->GetFileSize(manifest_filename, &current_manifest_file_size);
if (!s.ok()) {
return s;
}
bool have_log_number = false;
bool have_prev_log_number = false;
bool have_next_file = false;
bool have_last_sequence = false;
uint64_t next_file = 0;
uint64_t last_sequence = 0;
uint64_t log_number = 0;
uint64_t previous_log_number = 0;
uint32_t max_column_family = 0;
std::unordered_map<uint32_t, BaseReferencedVersionBuilder*> builders;
// add default column family
auto default_cf_iter = cf_name_to_options.find(kDefaultColumnFamilyName);
if (default_cf_iter == cf_name_to_options.end()) {
2014-03-13 02:09:03 +01:00
return Status::InvalidArgument("Default column family not specified");
}
2014-03-13 02:09:03 +01:00
VersionEdit default_cf_edit;
default_cf_edit.AddColumnFamily(kDefaultColumnFamilyName);
2014-03-13 02:09:03 +01:00
default_cf_edit.SetColumnFamily(0);
ColumnFamilyData* default_cfd =
CreateColumnFamily(default_cf_iter->second, &default_cf_edit);
builders.insert({0, new BaseReferencedVersionBuilder(default_cfd)});
{
VersionSet::LogReporter reporter;
reporter.status = &s;
log::Reader reader(std::move(manifest_file), &reporter, true /*checksum*/,
0 /*initial_offset*/);
Slice record;
std::string scratch;
while (reader.ReadRecord(&record, &scratch) && s.ok()) {
VersionEdit edit;
s = edit.DecodeFrom(record);
if (!s.ok()) {
break;
}
// Not found means that user didn't supply that column
// family option AND we encountered column family add
// record. Once we encounter column family drop record,
// we will delete the column family from
// column_families_not_found.
bool cf_in_not_found =
column_families_not_found.find(edit.column_family_) !=
column_families_not_found.end();
// in builders means that user supplied that column family
// option AND that we encountered column family add record
bool cf_in_builders =
builders.find(edit.column_family_) != builders.end();
// they can't both be true
assert(!(cf_in_not_found && cf_in_builders));
ColumnFamilyData* cfd = nullptr;
if (edit.is_column_family_add_) {
if (cf_in_builders || cf_in_not_found) {
s = Status::Corruption(
"Manifest adding the same column family twice");
break;
}
auto cf_options = cf_name_to_options.find(edit.column_family_name_);
if (cf_options == cf_name_to_options.end()) {
2014-04-09 19:38:05 +02:00
column_families_not_found.insert(
{edit.column_family_, edit.column_family_name_});
} else {
cfd = CreateColumnFamily(cf_options->second, &edit);
builders.insert(
{edit.column_family_, new BaseReferencedVersionBuilder(cfd)});
}
} else if (edit.is_column_family_drop_) {
if (cf_in_builders) {
auto builder = builders.find(edit.column_family_);
assert(builder != builders.end());
delete builder->second;
builders.erase(builder);
cfd = column_family_set_->GetColumnFamily(edit.column_family_);
if (cfd->Unref()) {
delete cfd;
cfd = nullptr;
} else {
// who else can have reference to cfd!?
assert(false);
}
} else if (cf_in_not_found) {
column_families_not_found.erase(edit.column_family_);
} else {
s = Status::Corruption(
"Manifest - dropping non-existing column family");
break;
}
} else if (!cf_in_not_found) {
if (!cf_in_builders) {
s = Status::Corruption(
"Manifest record referencing unknown column family");
break;
}
cfd = column_family_set_->GetColumnFamily(edit.column_family_);
// this should never happen since cf_in_builders is true
assert(cfd != nullptr);
if (edit.max_level_ >= cfd->current()->storage_info()->num_levels()) {
s = Status::InvalidArgument(
"db has more levels than options.num_levels");
break;
}
// if it is not column family add or column family drop,
// then it's a file add/delete, which should be forwarded
// to builder
auto builder = builders.find(edit.column_family_);
assert(builder != builders.end());
builder->second->version_builder()->Apply(&edit);
}
if (cfd != nullptr) {
if (edit.has_log_number_) {
2014-03-14 21:11:41 +01:00
if (cfd->GetLogNumber() > edit.log_number_) {
Log(InfoLogLevel::WARN_LEVEL, db_options_->info_log,
"MANIFEST corruption detected, but ignored - Log numbers in "
"records NOT monotonically increasing");
} else {
cfd->SetLogNumber(edit.log_number_);
have_log_number = true;
2014-03-14 21:11:41 +01:00
}
}
if (edit.has_comparator_ &&
edit.comparator_ != cfd->user_comparator()->Name()) {
s = Status::InvalidArgument(
cfd->user_comparator()->Name(),
"does not match existing comparator " + edit.comparator_);
break;
}
}
if (edit.has_prev_log_number_) {
previous_log_number = edit.prev_log_number_;
have_prev_log_number = true;
}
if (edit.has_next_file_number_) {
next_file = edit.next_file_number_;
have_next_file = true;
}
if (edit.has_max_column_family_) {
max_column_family = edit.max_column_family_;
}
if (edit.has_last_sequence_) {
last_sequence = edit.last_sequence_;
have_last_sequence = true;
}
}
}
if (s.ok()) {
if (!have_next_file) {
s = Status::Corruption("no meta-nextfile entry in descriptor");
} else if (!have_log_number) {
s = Status::Corruption("no meta-lognumber entry in descriptor");
} else if (!have_last_sequence) {
s = Status::Corruption("no last-sequence-number entry in descriptor");
}
if (!have_prev_log_number) {
previous_log_number = 0;
}
column_family_set_->UpdateMaxColumnFamily(max_column_family);
MarkFileNumberUsedDuringRecovery(previous_log_number);
MarkFileNumberUsedDuringRecovery(log_number);
}
// there were some column families in the MANIFEST that weren't specified
// in the argument. This is OK in read_only mode
if (read_only == false && !column_families_not_found.empty()) {
std::string list_of_not_found;
2014-04-09 19:38:05 +02:00
for (const auto& cf : column_families_not_found) {
list_of_not_found += ", " + cf.second;
}
list_of_not_found = list_of_not_found.substr(2);
s = Status::InvalidArgument(
2014-04-09 19:38:05 +02:00
"You have to open all column families. Column families not opened: " +
list_of_not_found);
}
if (s.ok()) {
for (auto cfd : *column_family_set_) {
auto builders_iter = builders.find(cfd->GetID());
assert(builders_iter != builders.end());
auto* builder = builders_iter->second->version_builder();
if (db_options_->max_open_files == -1) {
// unlimited table cache. Pre-load table handle now.
// Need to do it out of the mutex.
builder->LoadTableHandlers();
}
Version* v = new Version(cfd, this, current_version_number_++);
builder->SaveTo(v->storage_info());
Prevent segfault because SizeUnderCompaction was called without any locks. Summary: SizeBeingCompacted was called without any lock protection. This causes crashes, especially when running db_bench with value_size=128K. The fix is to compute SizeUnderCompaction while holding the mutex and passing in these values into the call to Finalize. (gdb) where #4 leveldb::VersionSet::SizeBeingCompacted (this=this@entry=0x7f0b490931c0, level=level@entry=4) at db/version_set.cc:1827 #5 0x000000000043a3c8 in leveldb::VersionSet::Finalize (this=this@entry=0x7f0b490931c0, v=v@entry=0x7f0b3b86b480) at db/version_set.cc:1420 #6 0x00000000004418d1 in leveldb::VersionSet::LogAndApply (this=0x7f0b490931c0, edit=0x7f0b3dc8c200, mu=0x7f0b490835b0, new_descriptor_log=<optimized out>) at db/version_set.cc:1016 #7 0x00000000004222b2 in leveldb::DBImpl::InstallCompactionResults (this=this@entry=0x7f0b49083400, compact=compact@entry=0x7f0b2b8330f0) at db/db_impl.cc:1473 #8 0x0000000000426027 in leveldb::DBImpl::DoCompactionWork (this=this@entry=0x7f0b49083400, compact=compact@entry=0x7f0b2b8330f0) at db/db_impl.cc:1757 #9 0x0000000000426690 in leveldb::DBImpl::BackgroundCompaction (this=this@entry=0x7f0b49083400, madeProgress=madeProgress@entry=0x7f0b41bf2d1e, deletion_state=...) at db/db_impl.cc:1268 #10 0x0000000000428f42 in leveldb::DBImpl::BackgroundCall (this=0x7f0b49083400) at db/db_impl.cc:1170 #11 0x000000000045348e in BGThread (this=0x7f0b49023100) at util/env_posix.cc:941 #12 leveldb::(anonymous namespace)::PosixEnv::BGThreadWrapper (arg=0x7f0b49023100) at util/env_posix.cc:874 #13 0x00007f0b4a7cf10d in start_thread (arg=0x7f0b41bf3700) at pthread_create.c:301 #14 0x00007f0b49b4b11d in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:115 Test Plan: make check I am running db_bench with a value size of 128K to see if the segfault is fixed. Reviewers: MarkCallaghan, sheki, emayanke Reviewed By: sheki CC: leveldb Differential Revision: https://reviews.facebook.net/D9279
2013-03-11 17:47:48 +01:00
// Install recovered version
std::vector<uint64_t> size_being_compacted(
v->storage_info()->num_levels() - 1);
cfd->compaction_picker()->SizeBeingCompacted(size_being_compacted);
v->PrepareApply(*cfd->GetLatestMutableCFOptions(), size_being_compacted);
AppendVersion(cfd, v);
}
Prevent segfault because SizeUnderCompaction was called without any locks. Summary: SizeBeingCompacted was called without any lock protection. This causes crashes, especially when running db_bench with value_size=128K. The fix is to compute SizeUnderCompaction while holding the mutex and passing in these values into the call to Finalize. (gdb) where #4 leveldb::VersionSet::SizeBeingCompacted (this=this@entry=0x7f0b490931c0, level=level@entry=4) at db/version_set.cc:1827 #5 0x000000000043a3c8 in leveldb::VersionSet::Finalize (this=this@entry=0x7f0b490931c0, v=v@entry=0x7f0b3b86b480) at db/version_set.cc:1420 #6 0x00000000004418d1 in leveldb::VersionSet::LogAndApply (this=0x7f0b490931c0, edit=0x7f0b3dc8c200, mu=0x7f0b490835b0, new_descriptor_log=<optimized out>) at db/version_set.cc:1016 #7 0x00000000004222b2 in leveldb::DBImpl::InstallCompactionResults (this=this@entry=0x7f0b49083400, compact=compact@entry=0x7f0b2b8330f0) at db/db_impl.cc:1473 #8 0x0000000000426027 in leveldb::DBImpl::DoCompactionWork (this=this@entry=0x7f0b49083400, compact=compact@entry=0x7f0b2b8330f0) at db/db_impl.cc:1757 #9 0x0000000000426690 in leveldb::DBImpl::BackgroundCompaction (this=this@entry=0x7f0b49083400, madeProgress=madeProgress@entry=0x7f0b41bf2d1e, deletion_state=...) at db/db_impl.cc:1268 #10 0x0000000000428f42 in leveldb::DBImpl::BackgroundCall (this=0x7f0b49083400) at db/db_impl.cc:1170 #11 0x000000000045348e in BGThread (this=0x7f0b49023100) at util/env_posix.cc:941 #12 leveldb::(anonymous namespace)::PosixEnv::BGThreadWrapper (arg=0x7f0b49023100) at util/env_posix.cc:874 #13 0x00007f0b4a7cf10d in start_thread (arg=0x7f0b41bf3700) at pthread_create.c:301 #14 0x00007f0b49b4b11d in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:115 Test Plan: make check I am running db_bench with a value size of 128K to see if the segfault is fixed. Reviewers: MarkCallaghan, sheki, emayanke Reviewed By: sheki CC: leveldb Differential Revision: https://reviews.facebook.net/D9279
2013-03-11 17:47:48 +01:00
manifest_file_size_ = current_manifest_file_size;
next_file_number_.store(next_file + 1);
last_sequence_ = last_sequence;
prev_log_number_ = previous_log_number;
Log(InfoLogLevel::INFO_LEVEL, db_options_->info_log,
"Recovered from manifest file:%s succeeded,"
"manifest_file_number is %lu, next_file_number is %lu, "
"last_sequence is %lu, log_number is %lu,"
"prev_log_number is %lu,"
"max_column_family is %u\n",
manifest_filename.c_str(), (unsigned long)manifest_file_number_,
(unsigned long)next_file_number_.load(), (unsigned long)last_sequence_,
(unsigned long)log_number, (unsigned long)prev_log_number_,
column_family_set_->GetMaxColumnFamily());
for (auto cfd : *column_family_set_) {
Log(InfoLogLevel::INFO_LEVEL, db_options_->info_log,
"Column family [%s] (ID %u), log number is %" PRIu64 "\n",
cfd->GetName().c_str(), cfd->GetID(), cfd->GetLogNumber());
}
}
for (auto builder : builders) {
delete builder.second;
}
return s;
}
Status VersionSet::ListColumnFamilies(std::vector<std::string>* column_families,
const std::string& dbname, Env* env) {
// these are just for performance reasons, not correcntes,
// so we're fine using the defaults
EnvOptions soptions;
// Read "CURRENT" file, which contains a pointer to the current manifest file
std::string current;
Status s = ReadFileToString(env, CurrentFileName(dbname), &current);
if (!s.ok()) {
return s;
}
if (current.empty() || current[current.size()-1] != '\n') {
return Status::Corruption("CURRENT file does not end with newline");
}
current.resize(current.size() - 1);
std::string dscname = dbname + "/" + current;
unique_ptr<SequentialFile> file;
s = env->NewSequentialFile(dscname, &file, soptions);
if (!s.ok()) {
return s;
}
std::map<uint32_t, std::string> column_family_names;
// default column family is always implicitly there
column_family_names.insert({0, kDefaultColumnFamilyName});
VersionSet::LogReporter reporter;
reporter.status = &s;
log::Reader reader(std::move(file), &reporter, true /*checksum*/,
0 /*initial_offset*/);
Slice record;
std::string scratch;
while (reader.ReadRecord(&record, &scratch) && s.ok()) {
VersionEdit edit;
s = edit.DecodeFrom(record);
if (!s.ok()) {
break;
}
if (edit.is_column_family_add_) {
if (column_family_names.find(edit.column_family_) !=
column_family_names.end()) {
s = Status::Corruption("Manifest adding the same column family twice");
break;
}
column_family_names.insert(
{edit.column_family_, edit.column_family_name_});
} else if (edit.is_column_family_drop_) {
if (column_family_names.find(edit.column_family_) ==
column_family_names.end()) {
s = Status::Corruption(
"Manifest - dropping non-existing column family");
break;
}
column_family_names.erase(edit.column_family_);
}
}
column_families->clear();
if (s.ok()) {
for (const auto& iter : column_family_names) {
column_families->push_back(iter.second);
Utility to dump manifest contents. Summary: ./manifest_dump --file=/tmp/dbbench/MANIFEST-000002 Output looks like manifest_file_number 30 next_file_number 31 last_sequence 388082 log_number 28 prev_log_number 0 --- level 0 --- --- level 1 --- --- level 2 --- 5:3244155['0000000000000000' @ 1 : 1 .. '0000000000028220' @ 28221 : 1] 7:3244177['0000000000028221' @ 28222 : 1 .. '0000000000056441' @ 56442 : 1] 9:3244156['0000000000056442' @ 56443 : 1 .. '0000000000084662' @ 84663 : 1] 11:3244178['0000000000084663' @ 84664 : 1 .. '0000000000112883' @ 112884 : 1] 13:3244158['0000000000112884' @ 112885 : 1 .. '0000000000141104' @ 141105 : 1] 15:3244176['0000000000141105' @ 141106 : 1 .. '0000000000169325' @ 169326 : 1] 17:3244156['0000000000169326' @ 169327 : 1 .. '0000000000197546' @ 197547 : 1] 19:3244178['0000000000197547' @ 197548 : 1 .. '0000000000225767' @ 225768 : 1] 21:3244155['0000000000225768' @ 225769 : 1 .. '0000000000253988' @ 253989 : 1] 23:3244179['0000000000253989' @ 253990 : 1 .. '0000000000282209' @ 282210 : 1] 25:3244157['0000000000282210' @ 282211 : 1 .. '0000000000310430' @ 310431 : 1] 27:3244176['0000000000310431' @ 310432 : 1 .. '0000000000338651' @ 338652 : 1] 29:3244156['0000000000338652' @ 338653 : 1 .. '0000000000366872' @ 366873 : 1] --- level 3 --- --- level 4 --- --- level 5 --- --- level 6 --- Test Plan: run on test directory created by dbbench Reviewers: heyongqiang Reviewed By: heyongqiang CC: hustliubo Differential Revision: https://reviews.facebook.net/D4743
2012-08-17 19:48:40 +02:00
}
}
return s;
}
Utility to dump manifest contents. Summary: ./manifest_dump --file=/tmp/dbbench/MANIFEST-000002 Output looks like manifest_file_number 30 next_file_number 31 last_sequence 388082 log_number 28 prev_log_number 0 --- level 0 --- --- level 1 --- --- level 2 --- 5:3244155['0000000000000000' @ 1 : 1 .. '0000000000028220' @ 28221 : 1] 7:3244177['0000000000028221' @ 28222 : 1 .. '0000000000056441' @ 56442 : 1] 9:3244156['0000000000056442' @ 56443 : 1 .. '0000000000084662' @ 84663 : 1] 11:3244178['0000000000084663' @ 84664 : 1 .. '0000000000112883' @ 112884 : 1] 13:3244158['0000000000112884' @ 112885 : 1 .. '0000000000141104' @ 141105 : 1] 15:3244176['0000000000141105' @ 141106 : 1 .. '0000000000169325' @ 169326 : 1] 17:3244156['0000000000169326' @ 169327 : 1 .. '0000000000197546' @ 197547 : 1] 19:3244178['0000000000197547' @ 197548 : 1 .. '0000000000225767' @ 225768 : 1] 21:3244155['0000000000225768' @ 225769 : 1 .. '0000000000253988' @ 253989 : 1] 23:3244179['0000000000253989' @ 253990 : 1 .. '0000000000282209' @ 282210 : 1] 25:3244157['0000000000282210' @ 282211 : 1 .. '0000000000310430' @ 310431 : 1] 27:3244176['0000000000310431' @ 310432 : 1 .. '0000000000338651' @ 338652 : 1] 29:3244156['0000000000338652' @ 338653 : 1 .. '0000000000366872' @ 366873 : 1] --- level 3 --- --- level 4 --- --- level 5 --- --- level 6 --- Test Plan: run on test directory created by dbbench Reviewers: heyongqiang Reviewed By: heyongqiang CC: hustliubo Differential Revision: https://reviews.facebook.net/D4743
2012-08-17 19:48:40 +02:00
#ifndef ROCKSDB_LITE
Status VersionSet::ReduceNumberOfLevels(const std::string& dbname,
const Options* options,
const EnvOptions& env_options,
int new_levels) {
if (new_levels <= 1) {
return Status::InvalidArgument(
"Number of levels needs to be bigger than 1");
}
ColumnFamilyOptions cf_options(*options);
std::shared_ptr<Cache> tc(NewLRUCache(
options->max_open_files - 10, options->table_cache_numshardbits,
options->table_cache_remove_scan_count_limit));
WriteController wc;
WriteBuffer wb(options->db_write_buffer_size);
VersionSet versions(dbname, options, env_options, tc.get(), &wb, &wc);
Status status;
std::vector<ColumnFamilyDescriptor> dummy;
ColumnFamilyDescriptor dummy_descriptor(kDefaultColumnFamilyName,
ColumnFamilyOptions(*options));
2014-02-01 04:44:48 +01:00
dummy.push_back(dummy_descriptor);
status = versions.Recover(dummy);
if (!status.ok()) {
return status;
}
Version* current_version =
versions.GetColumnFamilySet()->GetDefault()->current();
auto* vstorage = current_version->storage_info();
int current_levels = vstorage->num_levels();
if (current_levels <= new_levels) {
return Status::OK();
}
// Make sure there are file only on one level from
// (new_levels-1) to (current_levels-1)
int first_nonempty_level = -1;
int first_nonempty_level_filenum = 0;
for (int i = new_levels - 1; i < current_levels; i++) {
int file_num = vstorage->NumLevelFiles(i);
if (file_num != 0) {
if (first_nonempty_level < 0) {
first_nonempty_level = i;
first_nonempty_level_filenum = file_num;
} else {
char msg[255];
snprintf(msg, sizeof(msg),
"Found at least two levels containing files: "
"[%d:%d],[%d:%d].\n",
first_nonempty_level, first_nonempty_level_filenum, i,
file_num);
return Status::InvalidArgument(msg);
}
}
}
// we need to allocate an array with the old number of levels size to
// avoid SIGSEGV in WriteSnapshot()
// however, all levels bigger or equal to new_levels will be empty
std::vector<FileMetaData*>* new_files_list =
new std::vector<FileMetaData*>[current_levels];
for (int i = 0; i < new_levels - 1; i++) {
new_files_list[i] = vstorage->LevelFiles(i);
}
if (first_nonempty_level > 0) {
new_files_list[new_levels - 1] = vstorage->LevelFiles(first_nonempty_level);
}
delete[] vstorage -> files_;
vstorage->files_ = new_files_list;
vstorage->num_levels_ = new_levels;
MutableCFOptions mutable_cf_options(*options, ImmutableCFOptions(*options));
VersionEdit ve;
port::Mutex dummy_mutex;
MutexLock l(&dummy_mutex);
return versions.LogAndApply(
versions.GetColumnFamilySet()->GetDefault(),
mutable_cf_options, &ve, &dummy_mutex, nullptr, true);
}
Status VersionSet::DumpManifest(Options& options, std::string& dscname,
bool verbose, bool hex) {
Utility to dump manifest contents. Summary: ./manifest_dump --file=/tmp/dbbench/MANIFEST-000002 Output looks like manifest_file_number 30 next_file_number 31 last_sequence 388082 log_number 28 prev_log_number 0 --- level 0 --- --- level 1 --- --- level 2 --- 5:3244155['0000000000000000' @ 1 : 1 .. '0000000000028220' @ 28221 : 1] 7:3244177['0000000000028221' @ 28222 : 1 .. '0000000000056441' @ 56442 : 1] 9:3244156['0000000000056442' @ 56443 : 1 .. '0000000000084662' @ 84663 : 1] 11:3244178['0000000000084663' @ 84664 : 1 .. '0000000000112883' @ 112884 : 1] 13:3244158['0000000000112884' @ 112885 : 1 .. '0000000000141104' @ 141105 : 1] 15:3244176['0000000000141105' @ 141106 : 1 .. '0000000000169325' @ 169326 : 1] 17:3244156['0000000000169326' @ 169327 : 1 .. '0000000000197546' @ 197547 : 1] 19:3244178['0000000000197547' @ 197548 : 1 .. '0000000000225767' @ 225768 : 1] 21:3244155['0000000000225768' @ 225769 : 1 .. '0000000000253988' @ 253989 : 1] 23:3244179['0000000000253989' @ 253990 : 1 .. '0000000000282209' @ 282210 : 1] 25:3244157['0000000000282210' @ 282211 : 1 .. '0000000000310430' @ 310431 : 1] 27:3244176['0000000000310431' @ 310432 : 1 .. '0000000000338651' @ 338652 : 1] 29:3244156['0000000000338652' @ 338653 : 1 .. '0000000000366872' @ 366873 : 1] --- level 3 --- --- level 4 --- --- level 5 --- --- level 6 --- Test Plan: run on test directory created by dbbench Reviewers: heyongqiang Reviewed By: heyongqiang CC: hustliubo Differential Revision: https://reviews.facebook.net/D4743
2012-08-17 19:48:40 +02:00
// Open the specified manifest file.
unique_ptr<SequentialFile> file;
Status s = options.env->NewSequentialFile(dscname, &file, env_options_);
Utility to dump manifest contents. Summary: ./manifest_dump --file=/tmp/dbbench/MANIFEST-000002 Output looks like manifest_file_number 30 next_file_number 31 last_sequence 388082 log_number 28 prev_log_number 0 --- level 0 --- --- level 1 --- --- level 2 --- 5:3244155['0000000000000000' @ 1 : 1 .. '0000000000028220' @ 28221 : 1] 7:3244177['0000000000028221' @ 28222 : 1 .. '0000000000056441' @ 56442 : 1] 9:3244156['0000000000056442' @ 56443 : 1 .. '0000000000084662' @ 84663 : 1] 11:3244178['0000000000084663' @ 84664 : 1 .. '0000000000112883' @ 112884 : 1] 13:3244158['0000000000112884' @ 112885 : 1 .. '0000000000141104' @ 141105 : 1] 15:3244176['0000000000141105' @ 141106 : 1 .. '0000000000169325' @ 169326 : 1] 17:3244156['0000000000169326' @ 169327 : 1 .. '0000000000197546' @ 197547 : 1] 19:3244178['0000000000197547' @ 197548 : 1 .. '0000000000225767' @ 225768 : 1] 21:3244155['0000000000225768' @ 225769 : 1 .. '0000000000253988' @ 253989 : 1] 23:3244179['0000000000253989' @ 253990 : 1 .. '0000000000282209' @ 282210 : 1] 25:3244157['0000000000282210' @ 282211 : 1 .. '0000000000310430' @ 310431 : 1] 27:3244176['0000000000310431' @ 310432 : 1 .. '0000000000338651' @ 338652 : 1] 29:3244156['0000000000338652' @ 338653 : 1 .. '0000000000366872' @ 366873 : 1] --- level 3 --- --- level 4 --- --- level 5 --- --- level 6 --- Test Plan: run on test directory created by dbbench Reviewers: heyongqiang Reviewed By: heyongqiang CC: hustliubo Differential Revision: https://reviews.facebook.net/D4743
2012-08-17 19:48:40 +02:00
if (!s.ok()) {
return s;
}
bool have_prev_log_number = false;
bool have_next_file = false;
bool have_last_sequence = false;
uint64_t next_file = 0;
uint64_t last_sequence = 0;
uint64_t previous_log_number = 0;
int count = 0;
std::unordered_map<uint32_t, std::string> comparators;
std::unordered_map<uint32_t, BaseReferencedVersionBuilder*> builders;
// add default column family
VersionEdit default_cf_edit;
default_cf_edit.AddColumnFamily(kDefaultColumnFamilyName);
default_cf_edit.SetColumnFamily(0);
ColumnFamilyData* default_cfd =
CreateColumnFamily(ColumnFamilyOptions(options), &default_cf_edit);
builders.insert({0, new BaseReferencedVersionBuilder(default_cfd)});
Utility to dump manifest contents. Summary: ./manifest_dump --file=/tmp/dbbench/MANIFEST-000002 Output looks like manifest_file_number 30 next_file_number 31 last_sequence 388082 log_number 28 prev_log_number 0 --- level 0 --- --- level 1 --- --- level 2 --- 5:3244155['0000000000000000' @ 1 : 1 .. '0000000000028220' @ 28221 : 1] 7:3244177['0000000000028221' @ 28222 : 1 .. '0000000000056441' @ 56442 : 1] 9:3244156['0000000000056442' @ 56443 : 1 .. '0000000000084662' @ 84663 : 1] 11:3244178['0000000000084663' @ 84664 : 1 .. '0000000000112883' @ 112884 : 1] 13:3244158['0000000000112884' @ 112885 : 1 .. '0000000000141104' @ 141105 : 1] 15:3244176['0000000000141105' @ 141106 : 1 .. '0000000000169325' @ 169326 : 1] 17:3244156['0000000000169326' @ 169327 : 1 .. '0000000000197546' @ 197547 : 1] 19:3244178['0000000000197547' @ 197548 : 1 .. '0000000000225767' @ 225768 : 1] 21:3244155['0000000000225768' @ 225769 : 1 .. '0000000000253988' @ 253989 : 1] 23:3244179['0000000000253989' @ 253990 : 1 .. '0000000000282209' @ 282210 : 1] 25:3244157['0000000000282210' @ 282211 : 1 .. '0000000000310430' @ 310431 : 1] 27:3244176['0000000000310431' @ 310432 : 1 .. '0000000000338651' @ 338652 : 1] 29:3244156['0000000000338652' @ 338653 : 1 .. '0000000000366872' @ 366873 : 1] --- level 3 --- --- level 4 --- --- level 5 --- --- level 6 --- Test Plan: run on test directory created by dbbench Reviewers: heyongqiang Reviewed By: heyongqiang CC: hustliubo Differential Revision: https://reviews.facebook.net/D4743
2012-08-17 19:48:40 +02:00
{
VersionSet::LogReporter reporter;
Utility to dump manifest contents. Summary: ./manifest_dump --file=/tmp/dbbench/MANIFEST-000002 Output looks like manifest_file_number 30 next_file_number 31 last_sequence 388082 log_number 28 prev_log_number 0 --- level 0 --- --- level 1 --- --- level 2 --- 5:3244155['0000000000000000' @ 1 : 1 .. '0000000000028220' @ 28221 : 1] 7:3244177['0000000000028221' @ 28222 : 1 .. '0000000000056441' @ 56442 : 1] 9:3244156['0000000000056442' @ 56443 : 1 .. '0000000000084662' @ 84663 : 1] 11:3244178['0000000000084663' @ 84664 : 1 .. '0000000000112883' @ 112884 : 1] 13:3244158['0000000000112884' @ 112885 : 1 .. '0000000000141104' @ 141105 : 1] 15:3244176['0000000000141105' @ 141106 : 1 .. '0000000000169325' @ 169326 : 1] 17:3244156['0000000000169326' @ 169327 : 1 .. '0000000000197546' @ 197547 : 1] 19:3244178['0000000000197547' @ 197548 : 1 .. '0000000000225767' @ 225768 : 1] 21:3244155['0000000000225768' @ 225769 : 1 .. '0000000000253988' @ 253989 : 1] 23:3244179['0000000000253989' @ 253990 : 1 .. '0000000000282209' @ 282210 : 1] 25:3244157['0000000000282210' @ 282211 : 1 .. '0000000000310430' @ 310431 : 1] 27:3244176['0000000000310431' @ 310432 : 1 .. '0000000000338651' @ 338652 : 1] 29:3244156['0000000000338652' @ 338653 : 1 .. '0000000000366872' @ 366873 : 1] --- level 3 --- --- level 4 --- --- level 5 --- --- level 6 --- Test Plan: run on test directory created by dbbench Reviewers: heyongqiang Reviewed By: heyongqiang CC: hustliubo Differential Revision: https://reviews.facebook.net/D4743
2012-08-17 19:48:40 +02:00
reporter.status = &s;
log::Reader reader(std::move(file), &reporter, true/*checksum*/,
0/*initial_offset*/);
Utility to dump manifest contents. Summary: ./manifest_dump --file=/tmp/dbbench/MANIFEST-000002 Output looks like manifest_file_number 30 next_file_number 31 last_sequence 388082 log_number 28 prev_log_number 0 --- level 0 --- --- level 1 --- --- level 2 --- 5:3244155['0000000000000000' @ 1 : 1 .. '0000000000028220' @ 28221 : 1] 7:3244177['0000000000028221' @ 28222 : 1 .. '0000000000056441' @ 56442 : 1] 9:3244156['0000000000056442' @ 56443 : 1 .. '0000000000084662' @ 84663 : 1] 11:3244178['0000000000084663' @ 84664 : 1 .. '0000000000112883' @ 112884 : 1] 13:3244158['0000000000112884' @ 112885 : 1 .. '0000000000141104' @ 141105 : 1] 15:3244176['0000000000141105' @ 141106 : 1 .. '0000000000169325' @ 169326 : 1] 17:3244156['0000000000169326' @ 169327 : 1 .. '0000000000197546' @ 197547 : 1] 19:3244178['0000000000197547' @ 197548 : 1 .. '0000000000225767' @ 225768 : 1] 21:3244155['0000000000225768' @ 225769 : 1 .. '0000000000253988' @ 253989 : 1] 23:3244179['0000000000253989' @ 253990 : 1 .. '0000000000282209' @ 282210 : 1] 25:3244157['0000000000282210' @ 282211 : 1 .. '0000000000310430' @ 310431 : 1] 27:3244176['0000000000310431' @ 310432 : 1 .. '0000000000338651' @ 338652 : 1] 29:3244156['0000000000338652' @ 338653 : 1 .. '0000000000366872' @ 366873 : 1] --- level 3 --- --- level 4 --- --- level 5 --- --- level 6 --- Test Plan: run on test directory created by dbbench Reviewers: heyongqiang Reviewed By: heyongqiang CC: hustliubo Differential Revision: https://reviews.facebook.net/D4743
2012-08-17 19:48:40 +02:00
Slice record;
std::string scratch;
while (reader.ReadRecord(&record, &scratch) && s.ok()) {
VersionEdit edit;
Utility to dump manifest contents. Summary: ./manifest_dump --file=/tmp/dbbench/MANIFEST-000002 Output looks like manifest_file_number 30 next_file_number 31 last_sequence 388082 log_number 28 prev_log_number 0 --- level 0 --- --- level 1 --- --- level 2 --- 5:3244155['0000000000000000' @ 1 : 1 .. '0000000000028220' @ 28221 : 1] 7:3244177['0000000000028221' @ 28222 : 1 .. '0000000000056441' @ 56442 : 1] 9:3244156['0000000000056442' @ 56443 : 1 .. '0000000000084662' @ 84663 : 1] 11:3244178['0000000000084663' @ 84664 : 1 .. '0000000000112883' @ 112884 : 1] 13:3244158['0000000000112884' @ 112885 : 1 .. '0000000000141104' @ 141105 : 1] 15:3244176['0000000000141105' @ 141106 : 1 .. '0000000000169325' @ 169326 : 1] 17:3244156['0000000000169326' @ 169327 : 1 .. '0000000000197546' @ 197547 : 1] 19:3244178['0000000000197547' @ 197548 : 1 .. '0000000000225767' @ 225768 : 1] 21:3244155['0000000000225768' @ 225769 : 1 .. '0000000000253988' @ 253989 : 1] 23:3244179['0000000000253989' @ 253990 : 1 .. '0000000000282209' @ 282210 : 1] 25:3244157['0000000000282210' @ 282211 : 1 .. '0000000000310430' @ 310431 : 1] 27:3244176['0000000000310431' @ 310432 : 1 .. '0000000000338651' @ 338652 : 1] 29:3244156['0000000000338652' @ 338653 : 1 .. '0000000000366872' @ 366873 : 1] --- level 3 --- --- level 4 --- --- level 5 --- --- level 6 --- Test Plan: run on test directory created by dbbench Reviewers: heyongqiang Reviewed By: heyongqiang CC: hustliubo Differential Revision: https://reviews.facebook.net/D4743
2012-08-17 19:48:40 +02:00
s = edit.DecodeFrom(record);
if (!s.ok()) {
break;
Utility to dump manifest contents. Summary: ./manifest_dump --file=/tmp/dbbench/MANIFEST-000002 Output looks like manifest_file_number 30 next_file_number 31 last_sequence 388082 log_number 28 prev_log_number 0 --- level 0 --- --- level 1 --- --- level 2 --- 5:3244155['0000000000000000' @ 1 : 1 .. '0000000000028220' @ 28221 : 1] 7:3244177['0000000000028221' @ 28222 : 1 .. '0000000000056441' @ 56442 : 1] 9:3244156['0000000000056442' @ 56443 : 1 .. '0000000000084662' @ 84663 : 1] 11:3244178['0000000000084663' @ 84664 : 1 .. '0000000000112883' @ 112884 : 1] 13:3244158['0000000000112884' @ 112885 : 1 .. '0000000000141104' @ 141105 : 1] 15:3244176['0000000000141105' @ 141106 : 1 .. '0000000000169325' @ 169326 : 1] 17:3244156['0000000000169326' @ 169327 : 1 .. '0000000000197546' @ 197547 : 1] 19:3244178['0000000000197547' @ 197548 : 1 .. '0000000000225767' @ 225768 : 1] 21:3244155['0000000000225768' @ 225769 : 1 .. '0000000000253988' @ 253989 : 1] 23:3244179['0000000000253989' @ 253990 : 1 .. '0000000000282209' @ 282210 : 1] 25:3244157['0000000000282210' @ 282211 : 1 .. '0000000000310430' @ 310431 : 1] 27:3244176['0000000000310431' @ 310432 : 1 .. '0000000000338651' @ 338652 : 1] 29:3244156['0000000000338652' @ 338653 : 1 .. '0000000000366872' @ 366873 : 1] --- level 3 --- --- level 4 --- --- level 5 --- --- level 6 --- Test Plan: run on test directory created by dbbench Reviewers: heyongqiang Reviewed By: heyongqiang CC: hustliubo Differential Revision: https://reviews.facebook.net/D4743
2012-08-17 19:48:40 +02:00
}
// Write out each individual edit
if (verbose) {
printf("*************************Edit[%d] = %s\n",
count, edit.DebugString(hex).c_str());
}
count++;
bool cf_in_builders =
builders.find(edit.column_family_) != builders.end();
if (edit.has_comparator_) {
comparators.insert({edit.column_family_, edit.comparator_});
Utility to dump manifest contents. Summary: ./manifest_dump --file=/tmp/dbbench/MANIFEST-000002 Output looks like manifest_file_number 30 next_file_number 31 last_sequence 388082 log_number 28 prev_log_number 0 --- level 0 --- --- level 1 --- --- level 2 --- 5:3244155['0000000000000000' @ 1 : 1 .. '0000000000028220' @ 28221 : 1] 7:3244177['0000000000028221' @ 28222 : 1 .. '0000000000056441' @ 56442 : 1] 9:3244156['0000000000056442' @ 56443 : 1 .. '0000000000084662' @ 84663 : 1] 11:3244178['0000000000084663' @ 84664 : 1 .. '0000000000112883' @ 112884 : 1] 13:3244158['0000000000112884' @ 112885 : 1 .. '0000000000141104' @ 141105 : 1] 15:3244176['0000000000141105' @ 141106 : 1 .. '0000000000169325' @ 169326 : 1] 17:3244156['0000000000169326' @ 169327 : 1 .. '0000000000197546' @ 197547 : 1] 19:3244178['0000000000197547' @ 197548 : 1 .. '0000000000225767' @ 225768 : 1] 21:3244155['0000000000225768' @ 225769 : 1 .. '0000000000253988' @ 253989 : 1] 23:3244179['0000000000253989' @ 253990 : 1 .. '0000000000282209' @ 282210 : 1] 25:3244157['0000000000282210' @ 282211 : 1 .. '0000000000310430' @ 310431 : 1] 27:3244176['0000000000310431' @ 310432 : 1 .. '0000000000338651' @ 338652 : 1] 29:3244156['0000000000338652' @ 338653 : 1 .. '0000000000366872' @ 366873 : 1] --- level 3 --- --- level 4 --- --- level 5 --- --- level 6 --- Test Plan: run on test directory created by dbbench Reviewers: heyongqiang Reviewed By: heyongqiang CC: hustliubo Differential Revision: https://reviews.facebook.net/D4743
2012-08-17 19:48:40 +02:00
}
ColumnFamilyData* cfd = nullptr;
if (edit.is_column_family_add_) {
if (cf_in_builders) {
s = Status::Corruption(
"Manifest adding the same column family twice");
break;
}
cfd = CreateColumnFamily(ColumnFamilyOptions(options), &edit);
builders.insert(
{edit.column_family_, new BaseReferencedVersionBuilder(cfd)});
} else if (edit.is_column_family_drop_) {
if (!cf_in_builders) {
s = Status::Corruption(
"Manifest - dropping non-existing column family");
break;
}
auto builder_iter = builders.find(edit.column_family_);
delete builder_iter->second;
builders.erase(builder_iter);
comparators.erase(edit.column_family_);
cfd = column_family_set_->GetColumnFamily(edit.column_family_);
assert(cfd != nullptr);
cfd->Unref();
delete cfd;
cfd = nullptr;
} else {
if (!cf_in_builders) {
s = Status::Corruption(
"Manifest record referencing unknown column family");
break;
}
cfd = column_family_set_->GetColumnFamily(edit.column_family_);
// this should never happen since cf_in_builders is true
assert(cfd != nullptr);
// if it is not column family add or column family drop,
// then it's a file add/delete, which should be forwarded
// to builder
auto builder = builders.find(edit.column_family_);
assert(builder != builders.end());
builder->second->version_builder()->Apply(&edit);
Utility to dump manifest contents. Summary: ./manifest_dump --file=/tmp/dbbench/MANIFEST-000002 Output looks like manifest_file_number 30 next_file_number 31 last_sequence 388082 log_number 28 prev_log_number 0 --- level 0 --- --- level 1 --- --- level 2 --- 5:3244155['0000000000000000' @ 1 : 1 .. '0000000000028220' @ 28221 : 1] 7:3244177['0000000000028221' @ 28222 : 1 .. '0000000000056441' @ 56442 : 1] 9:3244156['0000000000056442' @ 56443 : 1 .. '0000000000084662' @ 84663 : 1] 11:3244178['0000000000084663' @ 84664 : 1 .. '0000000000112883' @ 112884 : 1] 13:3244158['0000000000112884' @ 112885 : 1 .. '0000000000141104' @ 141105 : 1] 15:3244176['0000000000141105' @ 141106 : 1 .. '0000000000169325' @ 169326 : 1] 17:3244156['0000000000169326' @ 169327 : 1 .. '0000000000197546' @ 197547 : 1] 19:3244178['0000000000197547' @ 197548 : 1 .. '0000000000225767' @ 225768 : 1] 21:3244155['0000000000225768' @ 225769 : 1 .. '0000000000253988' @ 253989 : 1] 23:3244179['0000000000253989' @ 253990 : 1 .. '0000000000282209' @ 282210 : 1] 25:3244157['0000000000282210' @ 282211 : 1 .. '0000000000310430' @ 310431 : 1] 27:3244176['0000000000310431' @ 310432 : 1 .. '0000000000338651' @ 338652 : 1] 29:3244156['0000000000338652' @ 338653 : 1 .. '0000000000366872' @ 366873 : 1] --- level 3 --- --- level 4 --- --- level 5 --- --- level 6 --- Test Plan: run on test directory created by dbbench Reviewers: heyongqiang Reviewed By: heyongqiang CC: hustliubo Differential Revision: https://reviews.facebook.net/D4743
2012-08-17 19:48:40 +02:00
}
if (cfd != nullptr && edit.has_log_number_) {
cfd->SetLogNumber(edit.log_number_);
}
Utility to dump manifest contents. Summary: ./manifest_dump --file=/tmp/dbbench/MANIFEST-000002 Output looks like manifest_file_number 30 next_file_number 31 last_sequence 388082 log_number 28 prev_log_number 0 --- level 0 --- --- level 1 --- --- level 2 --- 5:3244155['0000000000000000' @ 1 : 1 .. '0000000000028220' @ 28221 : 1] 7:3244177['0000000000028221' @ 28222 : 1 .. '0000000000056441' @ 56442 : 1] 9:3244156['0000000000056442' @ 56443 : 1 .. '0000000000084662' @ 84663 : 1] 11:3244178['0000000000084663' @ 84664 : 1 .. '0000000000112883' @ 112884 : 1] 13:3244158['0000000000112884' @ 112885 : 1 .. '0000000000141104' @ 141105 : 1] 15:3244176['0000000000141105' @ 141106 : 1 .. '0000000000169325' @ 169326 : 1] 17:3244156['0000000000169326' @ 169327 : 1 .. '0000000000197546' @ 197547 : 1] 19:3244178['0000000000197547' @ 197548 : 1 .. '0000000000225767' @ 225768 : 1] 21:3244155['0000000000225768' @ 225769 : 1 .. '0000000000253988' @ 253989 : 1] 23:3244179['0000000000253989' @ 253990 : 1 .. '0000000000282209' @ 282210 : 1] 25:3244157['0000000000282210' @ 282211 : 1 .. '0000000000310430' @ 310431 : 1] 27:3244176['0000000000310431' @ 310432 : 1 .. '0000000000338651' @ 338652 : 1] 29:3244156['0000000000338652' @ 338653 : 1 .. '0000000000366872' @ 366873 : 1] --- level 3 --- --- level 4 --- --- level 5 --- --- level 6 --- Test Plan: run on test directory created by dbbench Reviewers: heyongqiang Reviewed By: heyongqiang CC: hustliubo Differential Revision: https://reviews.facebook.net/D4743
2012-08-17 19:48:40 +02:00
if (edit.has_prev_log_number_) {
previous_log_number = edit.prev_log_number_;
Utility to dump manifest contents. Summary: ./manifest_dump --file=/tmp/dbbench/MANIFEST-000002 Output looks like manifest_file_number 30 next_file_number 31 last_sequence 388082 log_number 28 prev_log_number 0 --- level 0 --- --- level 1 --- --- level 2 --- 5:3244155['0000000000000000' @ 1 : 1 .. '0000000000028220' @ 28221 : 1] 7:3244177['0000000000028221' @ 28222 : 1 .. '0000000000056441' @ 56442 : 1] 9:3244156['0000000000056442' @ 56443 : 1 .. '0000000000084662' @ 84663 : 1] 11:3244178['0000000000084663' @ 84664 : 1 .. '0000000000112883' @ 112884 : 1] 13:3244158['0000000000112884' @ 112885 : 1 .. '0000000000141104' @ 141105 : 1] 15:3244176['0000000000141105' @ 141106 : 1 .. '0000000000169325' @ 169326 : 1] 17:3244156['0000000000169326' @ 169327 : 1 .. '0000000000197546' @ 197547 : 1] 19:3244178['0000000000197547' @ 197548 : 1 .. '0000000000225767' @ 225768 : 1] 21:3244155['0000000000225768' @ 225769 : 1 .. '0000000000253988' @ 253989 : 1] 23:3244179['0000000000253989' @ 253990 : 1 .. '0000000000282209' @ 282210 : 1] 25:3244157['0000000000282210' @ 282211 : 1 .. '0000000000310430' @ 310431 : 1] 27:3244176['0000000000310431' @ 310432 : 1 .. '0000000000338651' @ 338652 : 1] 29:3244156['0000000000338652' @ 338653 : 1 .. '0000000000366872' @ 366873 : 1] --- level 3 --- --- level 4 --- --- level 5 --- --- level 6 --- Test Plan: run on test directory created by dbbench Reviewers: heyongqiang Reviewed By: heyongqiang CC: hustliubo Differential Revision: https://reviews.facebook.net/D4743
2012-08-17 19:48:40 +02:00
have_prev_log_number = true;
}
if (edit.has_next_file_number_) {
next_file = edit.next_file_number_;
have_next_file = true;
}
if (edit.has_last_sequence_) {
last_sequence = edit.last_sequence_;
have_last_sequence = true;
}
if (edit.has_max_column_family_) {
column_family_set_->UpdateMaxColumnFamily(edit.max_column_family_);
}
Utility to dump manifest contents. Summary: ./manifest_dump --file=/tmp/dbbench/MANIFEST-000002 Output looks like manifest_file_number 30 next_file_number 31 last_sequence 388082 log_number 28 prev_log_number 0 --- level 0 --- --- level 1 --- --- level 2 --- 5:3244155['0000000000000000' @ 1 : 1 .. '0000000000028220' @ 28221 : 1] 7:3244177['0000000000028221' @ 28222 : 1 .. '0000000000056441' @ 56442 : 1] 9:3244156['0000000000056442' @ 56443 : 1 .. '0000000000084662' @ 84663 : 1] 11:3244178['0000000000084663' @ 84664 : 1 .. '0000000000112883' @ 112884 : 1] 13:3244158['0000000000112884' @ 112885 : 1 .. '0000000000141104' @ 141105 : 1] 15:3244176['0000000000141105' @ 141106 : 1 .. '0000000000169325' @ 169326 : 1] 17:3244156['0000000000169326' @ 169327 : 1 .. '0000000000197546' @ 197547 : 1] 19:3244178['0000000000197547' @ 197548 : 1 .. '0000000000225767' @ 225768 : 1] 21:3244155['0000000000225768' @ 225769 : 1 .. '0000000000253988' @ 253989 : 1] 23:3244179['0000000000253989' @ 253990 : 1 .. '0000000000282209' @ 282210 : 1] 25:3244157['0000000000282210' @ 282211 : 1 .. '0000000000310430' @ 310431 : 1] 27:3244176['0000000000310431' @ 310432 : 1 .. '0000000000338651' @ 338652 : 1] 29:3244156['0000000000338652' @ 338653 : 1 .. '0000000000366872' @ 366873 : 1] --- level 3 --- --- level 4 --- --- level 5 --- --- level 6 --- Test Plan: run on test directory created by dbbench Reviewers: heyongqiang Reviewed By: heyongqiang CC: hustliubo Differential Revision: https://reviews.facebook.net/D4743
2012-08-17 19:48:40 +02:00
}
}
file.reset();
Utility to dump manifest contents. Summary: ./manifest_dump --file=/tmp/dbbench/MANIFEST-000002 Output looks like manifest_file_number 30 next_file_number 31 last_sequence 388082 log_number 28 prev_log_number 0 --- level 0 --- --- level 1 --- --- level 2 --- 5:3244155['0000000000000000' @ 1 : 1 .. '0000000000028220' @ 28221 : 1] 7:3244177['0000000000028221' @ 28222 : 1 .. '0000000000056441' @ 56442 : 1] 9:3244156['0000000000056442' @ 56443 : 1 .. '0000000000084662' @ 84663 : 1] 11:3244178['0000000000084663' @ 84664 : 1 .. '0000000000112883' @ 112884 : 1] 13:3244158['0000000000112884' @ 112885 : 1 .. '0000000000141104' @ 141105 : 1] 15:3244176['0000000000141105' @ 141106 : 1 .. '0000000000169325' @ 169326 : 1] 17:3244156['0000000000169326' @ 169327 : 1 .. '0000000000197546' @ 197547 : 1] 19:3244178['0000000000197547' @ 197548 : 1 .. '0000000000225767' @ 225768 : 1] 21:3244155['0000000000225768' @ 225769 : 1 .. '0000000000253988' @ 253989 : 1] 23:3244179['0000000000253989' @ 253990 : 1 .. '0000000000282209' @ 282210 : 1] 25:3244157['0000000000282210' @ 282211 : 1 .. '0000000000310430' @ 310431 : 1] 27:3244176['0000000000310431' @ 310432 : 1 .. '0000000000338651' @ 338652 : 1] 29:3244156['0000000000338652' @ 338653 : 1 .. '0000000000366872' @ 366873 : 1] --- level 3 --- --- level 4 --- --- level 5 --- --- level 6 --- Test Plan: run on test directory created by dbbench Reviewers: heyongqiang Reviewed By: heyongqiang CC: hustliubo Differential Revision: https://reviews.facebook.net/D4743
2012-08-17 19:48:40 +02:00
if (s.ok()) {
if (!have_next_file) {
s = Status::Corruption("no meta-nextfile entry in descriptor");
printf("no meta-nextfile entry in descriptor");
} else if (!have_last_sequence) {
printf("no last-sequence-number entry in descriptor");
s = Status::Corruption("no last-sequence-number entry in descriptor");
}
if (!have_prev_log_number) {
previous_log_number = 0;
Utility to dump manifest contents. Summary: ./manifest_dump --file=/tmp/dbbench/MANIFEST-000002 Output looks like manifest_file_number 30 next_file_number 31 last_sequence 388082 log_number 28 prev_log_number 0 --- level 0 --- --- level 1 --- --- level 2 --- 5:3244155['0000000000000000' @ 1 : 1 .. '0000000000028220' @ 28221 : 1] 7:3244177['0000000000028221' @ 28222 : 1 .. '0000000000056441' @ 56442 : 1] 9:3244156['0000000000056442' @ 56443 : 1 .. '0000000000084662' @ 84663 : 1] 11:3244178['0000000000084663' @ 84664 : 1 .. '0000000000112883' @ 112884 : 1] 13:3244158['0000000000112884' @ 112885 : 1 .. '0000000000141104' @ 141105 : 1] 15:3244176['0000000000141105' @ 141106 : 1 .. '0000000000169325' @ 169326 : 1] 17:3244156['0000000000169326' @ 169327 : 1 .. '0000000000197546' @ 197547 : 1] 19:3244178['0000000000197547' @ 197548 : 1 .. '0000000000225767' @ 225768 : 1] 21:3244155['0000000000225768' @ 225769 : 1 .. '0000000000253988' @ 253989 : 1] 23:3244179['0000000000253989' @ 253990 : 1 .. '0000000000282209' @ 282210 : 1] 25:3244157['0000000000282210' @ 282211 : 1 .. '0000000000310430' @ 310431 : 1] 27:3244176['0000000000310431' @ 310432 : 1 .. '0000000000338651' @ 338652 : 1] 29:3244156['0000000000338652' @ 338653 : 1 .. '0000000000366872' @ 366873 : 1] --- level 3 --- --- level 4 --- --- level 5 --- --- level 6 --- Test Plan: run on test directory created by dbbench Reviewers: heyongqiang Reviewed By: heyongqiang CC: hustliubo Differential Revision: https://reviews.facebook.net/D4743
2012-08-17 19:48:40 +02:00
}
}
if (s.ok()) {
for (auto cfd : *column_family_set_) {
auto builders_iter = builders.find(cfd->GetID());
assert(builders_iter != builders.end());
auto builder = builders_iter->second->version_builder();
Version* v = new Version(cfd, this, current_version_number_++);
builder->SaveTo(v->storage_info());
std::vector<uint64_t> size_being_compacted(
v->storage_info()->num_levels() - 1);
2014-03-18 17:45:52 +01:00
cfd->compaction_picker()->SizeBeingCompacted(size_being_compacted);
v->PrepareApply(*cfd->GetLatestMutableCFOptions(), size_being_compacted);
printf("--------------- Column family \"%s\" (ID %u) --------------\n",
cfd->GetName().c_str(), (unsigned int)cfd->GetID());
printf("log number: %lu\n", (unsigned long)cfd->GetLogNumber());
auto comparator = comparators.find(cfd->GetID());
if (comparator != comparators.end()) {
printf("comparator: %s\n", comparator->second.c_str());
} else {
printf("comparator: <NO COMPARATOR>\n");
}
printf("%s \n", v->DebugString(hex).c_str());
delete v;
}
Prevent segfault because SizeUnderCompaction was called without any locks. Summary: SizeBeingCompacted was called without any lock protection. This causes crashes, especially when running db_bench with value_size=128K. The fix is to compute SizeUnderCompaction while holding the mutex and passing in these values into the call to Finalize. (gdb) where #4 leveldb::VersionSet::SizeBeingCompacted (this=this@entry=0x7f0b490931c0, level=level@entry=4) at db/version_set.cc:1827 #5 0x000000000043a3c8 in leveldb::VersionSet::Finalize (this=this@entry=0x7f0b490931c0, v=v@entry=0x7f0b3b86b480) at db/version_set.cc:1420 #6 0x00000000004418d1 in leveldb::VersionSet::LogAndApply (this=0x7f0b490931c0, edit=0x7f0b3dc8c200, mu=0x7f0b490835b0, new_descriptor_log=<optimized out>) at db/version_set.cc:1016 #7 0x00000000004222b2 in leveldb::DBImpl::InstallCompactionResults (this=this@entry=0x7f0b49083400, compact=compact@entry=0x7f0b2b8330f0) at db/db_impl.cc:1473 #8 0x0000000000426027 in leveldb::DBImpl::DoCompactionWork (this=this@entry=0x7f0b49083400, compact=compact@entry=0x7f0b2b8330f0) at db/db_impl.cc:1757 #9 0x0000000000426690 in leveldb::DBImpl::BackgroundCompaction (this=this@entry=0x7f0b49083400, madeProgress=madeProgress@entry=0x7f0b41bf2d1e, deletion_state=...) at db/db_impl.cc:1268 #10 0x0000000000428f42 in leveldb::DBImpl::BackgroundCall (this=0x7f0b49083400) at db/db_impl.cc:1170 #11 0x000000000045348e in BGThread (this=0x7f0b49023100) at util/env_posix.cc:941 #12 leveldb::(anonymous namespace)::PosixEnv::BGThreadWrapper (arg=0x7f0b49023100) at util/env_posix.cc:874 #13 0x00007f0b4a7cf10d in start_thread (arg=0x7f0b41bf3700) at pthread_create.c:301 #14 0x00007f0b49b4b11d in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:115 Test Plan: make check I am running db_bench with a value size of 128K to see if the segfault is fixed. Reviewers: MarkCallaghan, sheki, emayanke Reviewed By: sheki CC: leveldb Differential Revision: https://reviews.facebook.net/D9279
2013-03-11 17:47:48 +01:00
// Free builders
for (auto& builder : builders) {
delete builder.second;
}
next_file_number_.store(next_file + 1);
Utility to dump manifest contents. Summary: ./manifest_dump --file=/tmp/dbbench/MANIFEST-000002 Output looks like manifest_file_number 30 next_file_number 31 last_sequence 388082 log_number 28 prev_log_number 0 --- level 0 --- --- level 1 --- --- level 2 --- 5:3244155['0000000000000000' @ 1 : 1 .. '0000000000028220' @ 28221 : 1] 7:3244177['0000000000028221' @ 28222 : 1 .. '0000000000056441' @ 56442 : 1] 9:3244156['0000000000056442' @ 56443 : 1 .. '0000000000084662' @ 84663 : 1] 11:3244178['0000000000084663' @ 84664 : 1 .. '0000000000112883' @ 112884 : 1] 13:3244158['0000000000112884' @ 112885 : 1 .. '0000000000141104' @ 141105 : 1] 15:3244176['0000000000141105' @ 141106 : 1 .. '0000000000169325' @ 169326 : 1] 17:3244156['0000000000169326' @ 169327 : 1 .. '0000000000197546' @ 197547 : 1] 19:3244178['0000000000197547' @ 197548 : 1 .. '0000000000225767' @ 225768 : 1] 21:3244155['0000000000225768' @ 225769 : 1 .. '0000000000253988' @ 253989 : 1] 23:3244179['0000000000253989' @ 253990 : 1 .. '0000000000282209' @ 282210 : 1] 25:3244157['0000000000282210' @ 282211 : 1 .. '0000000000310430' @ 310431 : 1] 27:3244176['0000000000310431' @ 310432 : 1 .. '0000000000338651' @ 338652 : 1] 29:3244156['0000000000338652' @ 338653 : 1 .. '0000000000366872' @ 366873 : 1] --- level 3 --- --- level 4 --- --- level 5 --- --- level 6 --- Test Plan: run on test directory created by dbbench Reviewers: heyongqiang Reviewed By: heyongqiang CC: hustliubo Differential Revision: https://reviews.facebook.net/D4743
2012-08-17 19:48:40 +02:00
last_sequence_ = last_sequence;
prev_log_number_ = previous_log_number;
Utility to dump manifest contents. Summary: ./manifest_dump --file=/tmp/dbbench/MANIFEST-000002 Output looks like manifest_file_number 30 next_file_number 31 last_sequence 388082 log_number 28 prev_log_number 0 --- level 0 --- --- level 1 --- --- level 2 --- 5:3244155['0000000000000000' @ 1 : 1 .. '0000000000028220' @ 28221 : 1] 7:3244177['0000000000028221' @ 28222 : 1 .. '0000000000056441' @ 56442 : 1] 9:3244156['0000000000056442' @ 56443 : 1 .. '0000000000084662' @ 84663 : 1] 11:3244178['0000000000084663' @ 84664 : 1 .. '0000000000112883' @ 112884 : 1] 13:3244158['0000000000112884' @ 112885 : 1 .. '0000000000141104' @ 141105 : 1] 15:3244176['0000000000141105' @ 141106 : 1 .. '0000000000169325' @ 169326 : 1] 17:3244156['0000000000169326' @ 169327 : 1 .. '0000000000197546' @ 197547 : 1] 19:3244178['0000000000197547' @ 197548 : 1 .. '0000000000225767' @ 225768 : 1] 21:3244155['0000000000225768' @ 225769 : 1 .. '0000000000253988' @ 253989 : 1] 23:3244179['0000000000253989' @ 253990 : 1 .. '0000000000282209' @ 282210 : 1] 25:3244157['0000000000282210' @ 282211 : 1 .. '0000000000310430' @ 310431 : 1] 27:3244176['0000000000310431' @ 310432 : 1 .. '0000000000338651' @ 338652 : 1] 29:3244156['0000000000338652' @ 338653 : 1 .. '0000000000366872' @ 366873 : 1] --- level 3 --- --- level 4 --- --- level 5 --- --- level 6 --- Test Plan: run on test directory created by dbbench Reviewers: heyongqiang Reviewed By: heyongqiang CC: hustliubo Differential Revision: https://reviews.facebook.net/D4743
2012-08-17 19:48:40 +02:00
printf(
2014-03-12 18:52:32 +01:00
"next_file_number %lu last_sequence "
"%lu prev_log_number %lu max_column_family %u\n",
(unsigned long)next_file_number_.load(), (unsigned long)last_sequence,
(unsigned long)previous_log_number,
column_family_set_->GetMaxColumnFamily());
Utility to dump manifest contents. Summary: ./manifest_dump --file=/tmp/dbbench/MANIFEST-000002 Output looks like manifest_file_number 30 next_file_number 31 last_sequence 388082 log_number 28 prev_log_number 0 --- level 0 --- --- level 1 --- --- level 2 --- 5:3244155['0000000000000000' @ 1 : 1 .. '0000000000028220' @ 28221 : 1] 7:3244177['0000000000028221' @ 28222 : 1 .. '0000000000056441' @ 56442 : 1] 9:3244156['0000000000056442' @ 56443 : 1 .. '0000000000084662' @ 84663 : 1] 11:3244178['0000000000084663' @ 84664 : 1 .. '0000000000112883' @ 112884 : 1] 13:3244158['0000000000112884' @ 112885 : 1 .. '0000000000141104' @ 141105 : 1] 15:3244176['0000000000141105' @ 141106 : 1 .. '0000000000169325' @ 169326 : 1] 17:3244156['0000000000169326' @ 169327 : 1 .. '0000000000197546' @ 197547 : 1] 19:3244178['0000000000197547' @ 197548 : 1 .. '0000000000225767' @ 225768 : 1] 21:3244155['0000000000225768' @ 225769 : 1 .. '0000000000253988' @ 253989 : 1] 23:3244179['0000000000253989' @ 253990 : 1 .. '0000000000282209' @ 282210 : 1] 25:3244157['0000000000282210' @ 282211 : 1 .. '0000000000310430' @ 310431 : 1] 27:3244176['0000000000310431' @ 310432 : 1 .. '0000000000338651' @ 338652 : 1] 29:3244156['0000000000338652' @ 338653 : 1 .. '0000000000366872' @ 366873 : 1] --- level 3 --- --- level 4 --- --- level 5 --- --- level 6 --- Test Plan: run on test directory created by dbbench Reviewers: heyongqiang Reviewed By: heyongqiang CC: hustliubo Differential Revision: https://reviews.facebook.net/D4743
2012-08-17 19:48:40 +02:00
}
Utility to dump manifest contents. Summary: ./manifest_dump --file=/tmp/dbbench/MANIFEST-000002 Output looks like manifest_file_number 30 next_file_number 31 last_sequence 388082 log_number 28 prev_log_number 0 --- level 0 --- --- level 1 --- --- level 2 --- 5:3244155['0000000000000000' @ 1 : 1 .. '0000000000028220' @ 28221 : 1] 7:3244177['0000000000028221' @ 28222 : 1 .. '0000000000056441' @ 56442 : 1] 9:3244156['0000000000056442' @ 56443 : 1 .. '0000000000084662' @ 84663 : 1] 11:3244178['0000000000084663' @ 84664 : 1 .. '0000000000112883' @ 112884 : 1] 13:3244158['0000000000112884' @ 112885 : 1 .. '0000000000141104' @ 141105 : 1] 15:3244176['0000000000141105' @ 141106 : 1 .. '0000000000169325' @ 169326 : 1] 17:3244156['0000000000169326' @ 169327 : 1 .. '0000000000197546' @ 197547 : 1] 19:3244178['0000000000197547' @ 197548 : 1 .. '0000000000225767' @ 225768 : 1] 21:3244155['0000000000225768' @ 225769 : 1 .. '0000000000253988' @ 253989 : 1] 23:3244179['0000000000253989' @ 253990 : 1 .. '0000000000282209' @ 282210 : 1] 25:3244157['0000000000282210' @ 282211 : 1 .. '0000000000310430' @ 310431 : 1] 27:3244176['0000000000310431' @ 310432 : 1 .. '0000000000338651' @ 338652 : 1] 29:3244156['0000000000338652' @ 338653 : 1 .. '0000000000366872' @ 366873 : 1] --- level 3 --- --- level 4 --- --- level 5 --- --- level 6 --- Test Plan: run on test directory created by dbbench Reviewers: heyongqiang Reviewed By: heyongqiang CC: hustliubo Differential Revision: https://reviews.facebook.net/D4743
2012-08-17 19:48:40 +02:00
Utility to dump manifest contents. Summary: ./manifest_dump --file=/tmp/dbbench/MANIFEST-000002 Output looks like manifest_file_number 30 next_file_number 31 last_sequence 388082 log_number 28 prev_log_number 0 --- level 0 --- --- level 1 --- --- level 2 --- 5:3244155['0000000000000000' @ 1 : 1 .. '0000000000028220' @ 28221 : 1] 7:3244177['0000000000028221' @ 28222 : 1 .. '0000000000056441' @ 56442 : 1] 9:3244156['0000000000056442' @ 56443 : 1 .. '0000000000084662' @ 84663 : 1] 11:3244178['0000000000084663' @ 84664 : 1 .. '0000000000112883' @ 112884 : 1] 13:3244158['0000000000112884' @ 112885 : 1 .. '0000000000141104' @ 141105 : 1] 15:3244176['0000000000141105' @ 141106 : 1 .. '0000000000169325' @ 169326 : 1] 17:3244156['0000000000169326' @ 169327 : 1 .. '0000000000197546' @ 197547 : 1] 19:3244178['0000000000197547' @ 197548 : 1 .. '0000000000225767' @ 225768 : 1] 21:3244155['0000000000225768' @ 225769 : 1 .. '0000000000253988' @ 253989 : 1] 23:3244179['0000000000253989' @ 253990 : 1 .. '0000000000282209' @ 282210 : 1] 25:3244157['0000000000282210' @ 282211 : 1 .. '0000000000310430' @ 310431 : 1] 27:3244176['0000000000310431' @ 310432 : 1 .. '0000000000338651' @ 338652 : 1] 29:3244156['0000000000338652' @ 338653 : 1 .. '0000000000366872' @ 366873 : 1] --- level 3 --- --- level 4 --- --- level 5 --- --- level 6 --- Test Plan: run on test directory created by dbbench Reviewers: heyongqiang Reviewed By: heyongqiang CC: hustliubo Differential Revision: https://reviews.facebook.net/D4743
2012-08-17 19:48:40 +02:00
return s;
}
#endif // ROCKSDB_LITE
Utility to dump manifest contents. Summary: ./manifest_dump --file=/tmp/dbbench/MANIFEST-000002 Output looks like manifest_file_number 30 next_file_number 31 last_sequence 388082 log_number 28 prev_log_number 0 --- level 0 --- --- level 1 --- --- level 2 --- 5:3244155['0000000000000000' @ 1 : 1 .. '0000000000028220' @ 28221 : 1] 7:3244177['0000000000028221' @ 28222 : 1 .. '0000000000056441' @ 56442 : 1] 9:3244156['0000000000056442' @ 56443 : 1 .. '0000000000084662' @ 84663 : 1] 11:3244178['0000000000084663' @ 84664 : 1 .. '0000000000112883' @ 112884 : 1] 13:3244158['0000000000112884' @ 112885 : 1 .. '0000000000141104' @ 141105 : 1] 15:3244176['0000000000141105' @ 141106 : 1 .. '0000000000169325' @ 169326 : 1] 17:3244156['0000000000169326' @ 169327 : 1 .. '0000000000197546' @ 197547 : 1] 19:3244178['0000000000197547' @ 197548 : 1 .. '0000000000225767' @ 225768 : 1] 21:3244155['0000000000225768' @ 225769 : 1 .. '0000000000253988' @ 253989 : 1] 23:3244179['0000000000253989' @ 253990 : 1 .. '0000000000282209' @ 282210 : 1] 25:3244157['0000000000282210' @ 282211 : 1 .. '0000000000310430' @ 310431 : 1] 27:3244176['0000000000310431' @ 310432 : 1 .. '0000000000338651' @ 338652 : 1] 29:3244156['0000000000338652' @ 338653 : 1 .. '0000000000366872' @ 366873 : 1] --- level 3 --- --- level 4 --- --- level 5 --- --- level 6 --- Test Plan: run on test directory created by dbbench Reviewers: heyongqiang Reviewed By: heyongqiang CC: hustliubo Differential Revision: https://reviews.facebook.net/D4743
2012-08-17 19:48:40 +02:00
void VersionSet::MarkFileNumberUsedDuringRecovery(uint64_t number) {
// only called during recovery which is single threaded, so this works because
// there can't be concurrent calls
if (next_file_number_.load(std::memory_order_relaxed) <= number) {
next_file_number_.store(number + 1, std::memory_order_relaxed);
}
}
Status VersionSet::WriteSnapshot(log::Writer* log) {
// TODO: Break up into multiple records to reduce memory usage on recovery?
2014-03-13 02:09:03 +01:00
// WARNING: This method doesn't hold a mutex!!
// This is done without DB mutex lock held, but only within single-threaded
// LogAndApply. Column family manipulations can only happen within LogAndApply
2014-03-13 02:09:03 +01:00
// (the same single thread), so we're safe to iterate.
for (auto cfd : *column_family_set_) {
{
// Store column family info
VersionEdit edit;
if (cfd->GetID() != 0) {
// default column family is always there,
// no need to explicitly write it
edit.AddColumnFamily(cfd->GetName());
edit.SetColumnFamily(cfd->GetID());
}
edit.SetComparatorName(
cfd->internal_comparator().user_comparator()->Name());
std::string record;
if (!edit.EncodeTo(&record)) {
return Status::Corruption(
"Unable to Encode VersionEdit:" + edit.DebugString(true));
}
Status s = log->AddRecord(record);
if (!s.ok()) {
return s;
}
}
{
// Save files
VersionEdit edit;
edit.SetColumnFamily(cfd->GetID());
for (int level = 0; level < cfd->NumberLevels(); level++) {
for (const auto& f :
cfd->current()->storage_info()->LevelFiles(level)) {
edit.AddFile(level, f->fd.GetNumber(), f->fd.GetPathId(),
f->fd.GetFileSize(), f->smallest, f->largest,
f->smallest_seqno, f->largest_seqno);
}
}
edit.SetLogNumber(cfd->GetLogNumber());
std::string record;
if (!edit.EncodeTo(&record)) {
return Status::Corruption(
"Unable to Encode VersionEdit:" + edit.DebugString(true));
}
Status s = log->AddRecord(record);
if (!s.ok()) {
return s;
}
}
}
2014-03-13 02:09:03 +01:00
return Status::OK();
}
// Opens the mainfest file and reads all records
// till it finds the record we are looking for.
bool VersionSet::ManifestContains(uint64_t manifest_file_num,
const std::string& record) const {
std::string fname = DescriptorFileName(dbname_, manifest_file_num);
Log(InfoLogLevel::INFO_LEVEL, db_options_->info_log,
"ManifestContains: checking %s\n", fname.c_str());
unique_ptr<SequentialFile> file;
Status s = env_->NewSequentialFile(fname, &file, env_options_);
if (!s.ok()) {
Log(InfoLogLevel::INFO_LEVEL, db_options_->info_log,
"ManifestContains: %s\n", s.ToString().c_str());
Log(InfoLogLevel::INFO_LEVEL, db_options_->info_log,
"ManifestContains: is unable to reopen the manifest file %s",
fname.c_str());
return false;
}
log::Reader reader(std::move(file), nullptr, true/*checksum*/, 0);
Slice r;
std::string scratch;
bool result = false;
while (reader.ReadRecord(&r, &scratch)) {
if (r == Slice(record)) {
result = true;
break;
}
}
Log(InfoLogLevel::INFO_LEVEL, db_options_->info_log,
"ManifestContains: result = %d\n", result ? 1 : 0);
return result;
}
uint64_t VersionSet::ApproximateOffsetOf(Version* v, const InternalKey& ikey) {
uint64_t result = 0;
const auto* vstorage = v->storage_info();
for (int level = 0; level < vstorage->num_levels(); level++) {
const std::vector<FileMetaData*>& files = vstorage->LevelFiles(level);
for (size_t i = 0; i < files.size(); i++) {
if (v->cfd_->internal_comparator().Compare(files[i]->largest, ikey) <=
0) {
// Entire file is before "ikey", so just add the file size
result += files[i]->fd.GetFileSize();
} else if (v->cfd_->internal_comparator().Compare(files[i]->smallest,
ikey) > 0) {
// Entire file is after "ikey", so ignore
if (level > 0) {
// Files other than level 0 are sorted by meta->smallest, so
// no further files in this level will contain data for
// "ikey".
break;
}
} else {
// "ikey" falls in the range for this table. Add the
// approximate offset of "ikey" within the table.
TableReader* table_reader_ptr;
Iterator* iter = v->cfd_->table_cache()->NewIterator(
ReadOptions(), env_options_, v->cfd_->internal_comparator(),
files[i]->fd, &table_reader_ptr);
if (table_reader_ptr != nullptr) {
result += table_reader_ptr->ApproximateOffsetOf(ikey.Encode());
}
delete iter;
}
}
}
return result;
}
void VersionSet::AddLiveFiles(std::vector<FileDescriptor>* live_list) {
// pre-calculate space requirement
int64_t total_files = 0;
for (auto cfd : *column_family_set_) {
Version* dummy_versions = cfd->dummy_versions();
for (Version* v = dummy_versions->next_; v != dummy_versions;
v = v->next_) {
const auto* vstorage = v->storage_info();
for (int level = 0; level < vstorage->num_levels(); level++) {
total_files += vstorage->LevelFiles(level).size();
}
}
}
// just one time extension to the right size
live_list->reserve(live_list->size() + static_cast<size_t>(total_files));
for (auto cfd : *column_family_set_) {
auto* current = cfd->current();
bool found_current = false;
Version* dummy_versions = cfd->dummy_versions();
for (Version* v = dummy_versions->next_; v != dummy_versions;
v = v->next_) {
v->AddLiveFiles(live_list);
if (v == current) {
found_current = true;
}
}
if (!found_current && current != nullptr) {
// Should never happen unless it is a bug.
assert(false);
current->AddLiveFiles(live_list);
}
}
}
Iterator* VersionSet::MakeInputIterator(Compaction* c) {
auto cfd = c->column_family_data();
ReadOptions read_options;
read_options.verify_checksums =
c->mutable_cf_options()->verify_checksums_in_compaction;
read_options.fill_cache = false;
// Level-0 files have to be merged together. For other levels,
// we will make a concatenating iterator per level.
// TODO(opt): use concatenating iterator for level-0 if there is no overlap
const size_t space = (c->level() == 0 ? c->input_levels(0)->num_files +
c->num_input_levels() - 1
: c->num_input_levels());
Iterator** list = new Iterator* [space];
size_t num = 0;
for (size_t which = 0; which < c->num_input_levels(); which++) {
if (c->input_levels(which)->num_files != 0) {
if (c->level(which) == 0) {
const LevelFilesBrief* flevel = c->input_levels(which);
for (size_t i = 0; i < flevel->num_files; i++) {
list[num++] = cfd->table_cache()->NewIterator(
read_options, env_options_compactions_,
cfd->internal_comparator(), flevel->files[i].fd, nullptr,
true /* for compaction */);
}
} else {
// Create concatenating iterator for the files from this level
list[num++] = NewTwoLevelIterator(new LevelFileIteratorState(
cfd->table_cache(), read_options, env_options_,
cfd->internal_comparator(), true /* for_compaction */,
false /* prefix enabled */),
new LevelFileNumIterator(cfd->internal_comparator(),
c->input_levels(which)));
}
}
}
assert(num <= space);
Iterator* result =
NewMergingIterator(&c->column_family_data()->internal_comparator(), list,
static_cast<int>(num));
delete[] list;
return result;
}
// verify that the files listed in this compaction are present
// in the current version
bool VersionSet::VerifyCompactionFileConsistency(Compaction* c) {
#ifndef NDEBUG
Version* version = c->column_family_data()->current();
const VersionStorageInfo* vstorage = version->storage_info();
if (c->input_version() != version) {
Log(InfoLogLevel::INFO_LEVEL, db_options_->info_log,
"[%s] compaction output being applied to a different base version from"
" input version",
c->column_family_data()->GetName().c_str());
}
for (size_t input = 0; input < c->num_input_levels(); ++input) {
int level = c->level(input);
for (size_t i = 0; i < c->num_input_files(input); ++i) {
uint64_t number = c->input(input, i)->fd.GetNumber();
bool found = false;
for (unsigned int j = 0; j < vstorage->files_[level].size(); j++) {
FileMetaData* f = vstorage->files_[level][j];
if (f->fd.GetNumber() == number) {
found = true;
break;
}
}
if (!found) {
return false; // input files non existent in current version
}
}
}
#endif
return true; // everything good
}
Status VersionSet::GetMetadataForFile(uint64_t number, int* filelevel,
FileMetaData** meta,
ColumnFamilyData** cfd) {
for (auto cfd_iter : *column_family_set_) {
Version* version = cfd_iter->current();
const auto* vstorage = version->storage_info();
for (int level = 0; level < vstorage->num_levels(); level++) {
for (const auto& file : vstorage->LevelFiles(level)) {
if (file->fd.GetNumber() == number) {
*meta = file;
*filelevel = level;
*cfd = cfd_iter;
return Status::OK();
}
}
}
}
return Status::NotFound("File not present in any level");
}
void VersionSet::GetLiveFilesMetaData(std::vector<LiveFileMetaData>* metadata) {
for (auto cfd : *column_family_set_) {
for (int level = 0; level < cfd->NumberLevels(); level++) {
for (const auto& file :
cfd->current()->storage_info()->LevelFiles(level)) {
LiveFileMetaData filemetadata;
filemetadata.column_family_name = cfd->GetName();
uint32_t path_id = file->fd.GetPathId();
if (path_id < db_options_->db_paths.size()) {
filemetadata.db_path = db_options_->db_paths[path_id].path;
} else {
assert(!db_options_->db_paths.empty());
filemetadata.db_path = db_options_->db_paths.back().path;
}
filemetadata.name = MakeTableFileName("", file->fd.GetNumber());
filemetadata.level = level;
filemetadata.size = file->fd.GetFileSize();
filemetadata.smallestkey = file->smallest.user_key().ToString();
filemetadata.largestkey = file->largest.user_key().ToString();
filemetadata.smallest_seqno = file->smallest_seqno;
filemetadata.largest_seqno = file->largest_seqno;
metadata->push_back(filemetadata);
}
}
}
}
void VersionSet::GetObsoleteFiles(std::vector<FileMetaData*>* files) {
files->insert(files->end(), obsolete_files_.begin(), obsolete_files_.end());
obsolete_files_.clear();
}
ColumnFamilyData* VersionSet::CreateColumnFamily(
const ColumnFamilyOptions& cf_options, VersionEdit* edit) {
assert(edit->is_column_family_add_);
Version* dummy_versions = new Version(nullptr, this);
// Ref() dummy version once so that later we can call Unref() to delete it
// by avoiding calling "delete" explicitly (~Version is private)
dummy_versions->Ref();
auto new_cfd = column_family_set_->CreateColumnFamily(
edit->column_family_name_, edit->column_family_, dummy_versions,
cf_options);
2014-03-18 22:23:47 +01:00
Version* v = new Version(new_cfd, this, current_version_number_++);
AppendVersion(new_cfd, v);
// GetLatestMutableCFOptions() is safe here without mutex since the
// cfd is not available to client
new_cfd->CreateNewMemtable(*new_cfd->GetLatestMutableCFOptions());
2014-02-28 20:08:24 +01:00
new_cfd->SetLogNumber(edit->log_number_);
return new_cfd;
}
} // namespace rocksdb