2013-10-16 23:59:46 +02:00
|
|
|
// 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.
|
|
|
|
//
|
2011-03-18 23:37:00 +01:00
|
|
|
// 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.
|
|
|
|
//
|
|
|
|
// Thread-safe (provides internal synchronization)
|
|
|
|
|
2013-10-05 07:32:05 +02:00
|
|
|
#pragma once
|
2011-03-18 23:37:00 +01:00
|
|
|
#include <string>
|
2014-07-02 18:54:20 +02:00
|
|
|
#include <vector>
|
2011-03-18 23:37:00 +01:00
|
|
|
#include <stdint.h>
|
2014-01-28 06:58:46 +01:00
|
|
|
|
2011-03-18 23:37:00 +01:00
|
|
|
#include "db/dbformat.h"
|
|
|
|
#include "port/port.h"
|
2014-01-28 06:58:46 +01:00
|
|
|
#include "rocksdb/cache.h"
|
|
|
|
#include "rocksdb/env.h"
|
2013-10-29 01:54:09 +01:00
|
|
|
#include "rocksdb/table.h"
|
2014-09-05 01:18:36 +02:00
|
|
|
#include "rocksdb/options.h"
|
2014-01-28 06:58:46 +01:00
|
|
|
#include "table/table_reader.h"
|
2011-03-18 23:37:00 +01:00
|
|
|
|
2013-10-04 06:49:15 +02:00
|
|
|
namespace rocksdb {
|
2011-03-18 23:37:00 +01:00
|
|
|
|
|
|
|
class Env;
|
In DB::NewIterator(), try to allocate the whole iterator tree in an arena
Summary:
In this patch, try to allocate the whole iterator tree starting from DBIter from an arena
1. ArenaWrappedDBIter is created when serves as the entry point of an iterator tree, with an arena in it.
2. Add an option to create iterator from arena for following iterators: DBIter, MergingIterator, MemtableIterator, all mem table's iterators, all table reader's iterators and two level iterator.
3. MergeIteratorBuilder is created to incrementally build the tree of internal iterators. It is passed to mem table list and version set and add iterators to it.
Limitations:
(1) Only DB::NewIterator() without tailing uses the arena. Other cases, including readonly DB and compactions are still from malloc
(2) Two level iterator itself is allocated in arena, but not iterators inside it.
Test Plan: make all check
Reviewers: ljin, haobo
Reviewed By: haobo
Subscribers: leveldb, dhruba, yhchiang, igor
Differential Revision: https://reviews.facebook.net/D18513
2014-06-03 01:38:00 +02:00
|
|
|
class Arena;
|
2014-06-14 00:54:19 +02:00
|
|
|
struct FileDescriptor;
|
2014-09-29 20:09:09 +02:00
|
|
|
class GetContext;
|
Measure file read latency histogram per level
Summary: In internal stats, remember read latency histogram, if statistics is enabled. It can be retrieved from DB::GetProperty() with "rocksdb.dbstats" property, if it is enabled.
Test Plan: Manually run db_bench and prints out "rocksdb.dbstats" by hand and make sure it prints out as expected
Reviewers: igor, IslamAbdelRahman, rven, kradhakrishnan, anthony, yhchiang
Reviewed By: yhchiang
Subscribers: MarkCallaghan, leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D44193
2015-08-13 23:35:54 +02:00
|
|
|
class HistogramImpl;
|
2011-03-18 23:37:00 +01:00
|
|
|
|
|
|
|
class TableCache {
|
|
|
|
public:
|
2014-09-05 01:18:36 +02:00
|
|
|
TableCache(const ImmutableCFOptions& ioptions,
|
|
|
|
const EnvOptions& storage_options, Cache* cache);
|
2011-03-18 23:37:00 +01:00
|
|
|
~TableCache();
|
|
|
|
|
2011-03-28 22:43:44 +02:00
|
|
|
// Return an iterator for the specified file number (the corresponding
|
|
|
|
// file length must be exactly "file_size" bytes). If "tableptr" is
|
2013-03-01 03:04:58 +01:00
|
|
|
// non-nullptr, also sets "*tableptr" to point to the Table object
|
|
|
|
// underlying the returned iterator, or nullptr if no Table object underlies
|
2011-03-28 22:43:44 +02:00
|
|
|
// the returned iterator. The returned "*tableptr" object is owned by
|
|
|
|
// the cache and should not be deleted, and is valid for as long as the
|
|
|
|
// returned iterator is live.
|
2014-01-27 22:53:22 +01:00
|
|
|
Iterator* NewIterator(const ReadOptions& options, const EnvOptions& toptions,
|
|
|
|
const InternalKeyComparator& internal_comparator,
|
2014-06-14 00:54:19 +02:00
|
|
|
const FileDescriptor& file_fd,
|
2013-10-30 18:52:33 +01:00
|
|
|
TableReader** table_reader_ptr = nullptr,
|
Measure file read latency histogram per level
Summary: In internal stats, remember read latency histogram, if statistics is enabled. It can be retrieved from DB::GetProperty() with "rocksdb.dbstats" property, if it is enabled.
Test Plan: Manually run db_bench and prints out "rocksdb.dbstats" by hand and make sure it prints out as expected
Reviewers: igor, IslamAbdelRahman, rven, kradhakrishnan, anthony, yhchiang
Reviewed By: yhchiang
Subscribers: MarkCallaghan, leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D44193
2015-08-13 23:35:54 +02:00
|
|
|
HistogramImpl* file_read_hist = nullptr,
|
In DB::NewIterator(), try to allocate the whole iterator tree in an arena
Summary:
In this patch, try to allocate the whole iterator tree starting from DBIter from an arena
1. ArenaWrappedDBIter is created when serves as the entry point of an iterator tree, with an arena in it.
2. Add an option to create iterator from arena for following iterators: DBIter, MergingIterator, MemtableIterator, all mem table's iterators, all table reader's iterators and two level iterator.
3. MergeIteratorBuilder is created to incrementally build the tree of internal iterators. It is passed to mem table list and version set and add iterators to it.
Limitations:
(1) Only DB::NewIterator() without tailing uses the arena. Other cases, including readonly DB and compactions are still from malloc
(2) Two level iterator itself is allocated in arena, but not iterators inside it.
Test Plan: make all check
Reviewers: ljin, haobo
Reviewed By: haobo
Subscribers: leveldb, dhruba, yhchiang, igor
Differential Revision: https://reviews.facebook.net/D18513
2014-06-03 01:38:00 +02:00
|
|
|
bool for_compaction = false, Arena* arena = nullptr);
|
2011-03-18 23:37:00 +01:00
|
|
|
|
2012-04-17 17:36:46 +02:00
|
|
|
// If a seek to internal key "k" in specified file finds an entry,
|
2013-03-21 23:59:47 +01:00
|
|
|
// call (*handle_result)(arg, found_key, found_value) repeatedly until
|
|
|
|
// it returns false.
|
2012-04-17 17:36:46 +02:00
|
|
|
Status Get(const ReadOptions& options,
|
2014-01-27 22:53:22 +01:00
|
|
|
const InternalKeyComparator& internal_comparator,
|
2014-09-29 20:09:09 +02:00
|
|
|
const FileDescriptor& file_fd, const Slice& k,
|
Measure file read latency histogram per level
Summary: In internal stats, remember read latency histogram, if statistics is enabled. It can be retrieved from DB::GetProperty() with "rocksdb.dbstats" property, if it is enabled.
Test Plan: Manually run db_bench and prints out "rocksdb.dbstats" by hand and make sure it prints out as expected
Reviewers: igor, IslamAbdelRahman, rven, kradhakrishnan, anthony, yhchiang
Reviewed By: yhchiang
Subscribers: MarkCallaghan, leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D44193
2015-08-13 23:35:54 +02:00
|
|
|
GetContext* get_context, HistogramImpl* file_read_hist = nullptr);
|
2012-04-17 17:36:46 +02:00
|
|
|
|
2011-03-18 23:37:00 +01:00
|
|
|
// Evict any entry for the specified file number
|
[CF] Rethink table cache
Summary:
Adapting table cache to column families is interesting. We want table cache to be global LRU, so if some column families are use not as often as others, we want them to be evicted from cache. However, current TableCache object also constructs tables on its own. If table is not found in the cache, TableCache automatically creates new table. We want each column family to be able to specify different table factory.
To solve the problem, we still have a single LRU, but we provide the LRUCache object to TableCache on construction. We have one TableCache per column family, but the underyling cache is shared by all TableCache objects.
This allows us to have a global LRU, but still be able to support different table factories for different column families. Also, in the future it will also be able to support different directories for different column families.
Test Plan: make check
Reviewers: dhruba, haobo, kailiu, sdong
CC: leveldb
Differential Revision: https://reviews.facebook.net/D15915
2014-02-05 18:07:55 +01:00
|
|
|
static void Evict(Cache* cache, uint64_t file_number);
|
2011-03-18 23:37:00 +01:00
|
|
|
|
2015-08-21 07:33:25 +02:00
|
|
|
// Build a table reader
|
|
|
|
Status GetTableReader(const EnvOptions& env_options,
|
|
|
|
const InternalKeyComparator& internal_comparator,
|
|
|
|
const FileDescriptor& fd, bool advise_random_on_open,
|
|
|
|
bool record_read_stats, HistogramImpl* file_read_hist,
|
|
|
|
unique_ptr<TableReader>* table_reader);
|
|
|
|
|
2014-01-07 05:29:17 +01:00
|
|
|
// Find table reader
|
2014-01-27 22:53:22 +01:00
|
|
|
Status FindTable(const EnvOptions& toptions,
|
|
|
|
const InternalKeyComparator& internal_comparator,
|
2014-06-14 00:54:19 +02:00
|
|
|
const FileDescriptor& file_fd, Cache::Handle**,
|
Measure file read latency histogram per level
Summary: In internal stats, remember read latency histogram, if statistics is enabled. It can be retrieved from DB::GetProperty() with "rocksdb.dbstats" property, if it is enabled.
Test Plan: Manually run db_bench and prints out "rocksdb.dbstats" by hand and make sure it prints out as expected
Reviewers: igor, IslamAbdelRahman, rven, kradhakrishnan, anthony, yhchiang
Reviewed By: yhchiang
Subscribers: MarkCallaghan, leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D44193
2015-08-13 23:35:54 +02:00
|
|
|
const bool no_io = false, bool record_read_stats = true,
|
|
|
|
HistogramImpl* file_read_hist = nullptr);
|
2014-01-07 05:29:17 +01:00
|
|
|
|
|
|
|
// Get TableReader from a cache handle.
|
|
|
|
TableReader* GetTableReaderFromHandle(Cache::Handle* handle);
|
|
|
|
|
2014-02-14 01:28:21 +01:00
|
|
|
// Get the table properties of a given table.
|
|
|
|
// @no_io: indicates if we should load table to the cache if it is not present
|
|
|
|
// in table cache yet.
|
|
|
|
// @returns: `properties` will be reset on success. Please note that we will
|
|
|
|
// return Status::Incomplete() if table is not present in cache and
|
|
|
|
// we set `no_io` to be true.
|
|
|
|
Status GetTableProperties(const EnvOptions& toptions,
|
|
|
|
const InternalKeyComparator& internal_comparator,
|
2014-06-14 00:54:19 +02:00
|
|
|
const FileDescriptor& file_meta,
|
2014-02-14 01:28:21 +01:00
|
|
|
std::shared_ptr<const TableProperties>* properties,
|
|
|
|
bool no_io = false);
|
|
|
|
|
2014-08-05 20:27:34 +02:00
|
|
|
// Return total memory usage of the table reader of the file.
|
|
|
|
// 0 of table reader of the file is not loaded.
|
|
|
|
size_t GetMemoryUsageByTableReader(
|
|
|
|
const EnvOptions& toptions,
|
|
|
|
const InternalKeyComparator& internal_comparator,
|
|
|
|
const FileDescriptor& fd);
|
|
|
|
|
2014-01-07 05:29:17 +01:00
|
|
|
// Release the handle from a cache
|
|
|
|
void ReleaseHandle(Cache::Handle* handle);
|
|
|
|
|
2011-03-18 23:37:00 +01:00
|
|
|
private:
|
2014-09-05 01:18:36 +02:00
|
|
|
const ImmutableCFOptions& ioptions_;
|
|
|
|
const EnvOptions& env_options_;
|
[CF] Rethink table cache
Summary:
Adapting table cache to column families is interesting. We want table cache to be global LRU, so if some column families are use not as often as others, we want them to be evicted from cache. However, current TableCache object also constructs tables on its own. If table is not found in the cache, TableCache automatically creates new table. We want each column family to be able to specify different table factory.
To solve the problem, we still have a single LRU, but we provide the LRUCache object to TableCache on construction. We have one TableCache per column family, but the underyling cache is shared by all TableCache objects.
This allows us to have a global LRU, but still be able to support different table factories for different column families. Also, in the future it will also be able to support different directories for different column families.
Test Plan: make check
Reviewers: dhruba, haobo, kailiu, sdong
CC: leveldb
Differential Revision: https://reviews.facebook.net/D15915
2014-02-05 18:07:55 +01:00
|
|
|
Cache* const cache_;
|
2015-06-23 19:25:45 +02:00
|
|
|
std::string row_cache_id_;
|
2011-03-18 23:37:00 +01:00
|
|
|
};
|
|
|
|
|
2013-10-04 06:49:15 +02:00
|
|
|
} // namespace rocksdb
|