Summary:
= Major Changes =
* Add a new mem-table representation, HashCuckooRep, which is based cuckoo hash.
Cuckoo hash uses multiple hash functions. This allows each key to have multiple
possible locations in the mem-table.
- Put: When insert a key, it will try to find whether one of its possible
locations is vacant and store the key. If none of its possible
locations are available, then it will kick out a victim key and
store at that location. The kicked-out victim key will then be
stored at a vacant space of its possible locations or kick-out
another victim. In this diff, the kick-out path (known as
cuckoo-path) is found using BFS, which guarantees to be the shortest.
- Get: Simply tries all possible locations of a key --- this guarantees
worst-case constant time complexity.
- Time complexity: O(1) for Get, and average O(1) for Put if the
fullness of the mem-table is below 80%.
- Default using two hash functions, the number of hash functions used
by the cuckoo-hash may dynamically increase if it fails to find a
short-enough kick-out path.
- Currently, HashCuckooRep does not support iteration and snapshots,
as our current main purpose of this is to optimize point access.
= Minor Changes =
* Add IsSnapshotSupported() to DB to indicate whether the current DB
supports snapshots. If it returns false, then DB::GetSnapshot() will
always return nullptr.
Test Plan:
Run existing tests. Will develop a test specifically for cuckoo hash in
the next diff.
Reviewers: sdong, haobo
Reviewed By: sdong
CC: leveldb, dhruba, igor
Differential Revision: https://reviews.facebook.net/D16155
Summary:
ReadFirstRecord() reads the actual log file from disk on every call. This diff introduces a cache layer on top of ReadFirstRecord(), which should significantly speed up repeated calls to GetUpdatesSince().
I also cleaned up some stuff, but the whole TransactionLogIterator could use some refactoring, especially if we see increased usage.
Test Plan: make check
Reviewers: haobo, sdong, dhruba
Reviewed By: haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D18387
Summary:
also add an override option total_order_iteration if you want to use full
iterator with prefix_extractor
Test Plan: make all check
Reviewers: igor, haobo, sdong, yhchiang
Reviewed By: haobo
CC: leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D17805
Summary: As summary. Add two autovectors that get filled up in MakeRoomForWrite and they get deleted outside of mutex
Test Plan: make check
Reviewers: dhruba, haobo, ljin, sdong
Reviewed By: ljin
CC: leveldb
Differential Revision: https://reviews.facebook.net/D18249
Summary:
Now that we have column families involved, we need to add extra context to every log message. They now start with "[column family name] log message"
Also added some logging that I think would be useful, like level summary after every flush (I often needed that when going through the logs).
Test Plan: make check + ran db_bench to confirm I'm happy with log output
Reviewers: dhruba, haobo, ljin, yhchiang, sdong
Reviewed By: haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D18303
Summary:
One of our profilings shows that Version::Get() sometimes is slow when getting pointer of user comparators or other global objects. In this patch:
(1) we keep pointers of immutable objects in Version to avoid accesses them though option objects or cfd objects
(2) table_reader is directly cached in FileMetaData so that table cache don't have to go through handle first to fetch it
(3) If level 0 has less than 3 files, skip the filtering logic based on SST tables' key range. Smallest and largest key are stored in separated memory locations, which has potential cache misses
Test Plan: make all check
Reviewers: haobo, ljin
Reviewed By: haobo
CC: igor, yhchiang, nkg-, leveldb
Differential Revision: https://reviews.facebook.net/D17739
Summary: Current behavior of creating new DB is, if there is existing log files, we will go ahead and replay them on top of empty DB. This is a behavior that no user would expect. With this patch, we will fail the creation if a user creates a DB with existing log files.
Test Plan: make all check
Reviewers: haobo, igor, ljin
Reviewed By: haobo
CC: nkg-, yhchiang, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D17817
Summary:
Introducing RocksDBLite! Removes all the non-essential features and reduces the binary size. This effort should help our adoption on mobile.
Binary size when compiling for IOS (`TARGET_OS=IOS m static_lib`) is down to 9MB from 15MB (without stripping)
Test Plan: compiles :)
Reviewers: dhruba, haobo, ljin, sdong, yhchiang
Reviewed By: yhchiang
CC: leveldb
Differential Revision: https://reviews.facebook.net/D17835
Summary:
With multiple column families, especially when manual Flush is executed, we might roll the log file, although the current log file is empty (no data has been written to the log).
After the diff, we won't create new log file if current is empty.
Next, I will write an algorithm that will flush column families that reference old log files (i.e., that weren't flushed in a while)
Test Plan: Added an unit test. Confirmed that unit test failes in master
Reviewers: dhruba, haobo, ljin, sdong
Reviewed By: ljin
CC: leveldb
Differential Revision: https://reviews.facebook.net/D17631
Summary:
replace the super version acquisision in tailing itrator with thread
local
Test Plan: will post results
Reviewers: igor, haobo, sdong, yhchiang, dhruba
Reviewed By: igor
CC: leveldb
Differential Revision: https://reviews.facebook.net/D17757
Summary:
Similar to GetImp(), use SuperVersion from thread local instead of acquriing mutex.
I don't expect this change will make a dent on NewIterator() performance
because the bottleneck seems to be on the rest part of the API
Test Plan:
make asan_check
will post perf numbers
Reviewers: haobo, igor, sdong, dhruba, yhchiang
Reviewed By: sdong
CC: leveldb
Differential Revision: https://reviews.facebook.net/D17643
Summary: XCode for some reason injects `#define DEBUG 1` into our code, which makes compile fail because we use `DEBUG` keyword for other stuff. This diff fixes the issue by renaming `DEBUG` to `DEBUG_LEVEL`.
Test Plan: compiles
Reviewers: dhruba, haobo, sdong, yhchiang, ljin
Reviewed By: yhchiang
CC: leveldb
Differential Revision: https://reviews.facebook.net/D17709
Summary:
1. Polish IterKey a little bit.
2. Turn to use it in local parameter of current_user_key in DBImpl::ProcessKeyValueCompaction(). Our profile showing that DBImpl::ProcessKeyValueCompaction() has about 14% costs in std::string (the base including reading and writing data but excluding compaction filtering), which is higher than it should be. There are two std::string used in DBImpl::ProcessKeyValueCompaction(), compaction_filter_value and current_user_key and it's hard to distinguish the two.
Test Plan: make all check
Reviewers: haobo, ljin
Reviewed By: haobo
CC: igor, yhchiang, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D17613
Summary: When opening DB in read-only mode, client can choose to only specify a subset of column families ("default" column family can't be omitted, though)
Test Plan: added a unit test in column_family_test
Reviewers: haobo, sdong, ljin, dhruba
Reviewed By: haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D17565
Summary: ToString() is expensive. Profiling shows that most compaction threads are stuck in jemalloc, allocating a new string. This will help out a litte.
Test Plan: make check
Reviewers: haobo, danguo
Reviewed By: danguo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D17583
Summary: This will allow us to disable them completely for iOS or for better performance
Test Plan: will run make all check
Reviewers: igor, haobo, dhruba
Reviewed By: haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D17511
Summary: If a client specifies wal_dir with trailing '/', we will fail in deleting obsolete log files. See task #4083746
Test Plan: make check
Reviewers: haobo, sdong
Reviewed By: haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D17535
Summary: Our measurement shows that sometimes new log::Write's constructor can take hundreds of milliseconds. It's unclear why but just simply move it out of DB mutex.
Test Plan: make all check
Reviewers: haobo, ljin, igor
Reviewed By: haobo
CC: nkg-, yhchiang, leveldb
Differential Revision: https://reviews.facebook.net/D17487
Summary: Flushing log buffer earlier to avoid confusion of time holding the locks.
Test Plan: Should be safe as long as several related db test passes
Reviewers: haobo, igor, ljin
Reviewed By: igor
CC: nkg-, leveldb
Differential Revision: https://reviews.facebook.net/D17493
Summary: Move several some common logging still in DB mutex to log buffer.
Test Plan: make all check
Reviewers: haobo, igor, ljin, nkg-
Reviewed By: nkg-
CC: nkg-, yhchiang, leveldb
Differential Revision: https://reviews.facebook.net/D17439
Summary: This patch fixed a race condition where a log file is moved to archived dir in the middle of GetSortedWalFiles. Without the fix, the log file would be missed in the result, which leads to transaction log iterator gap. A test utility SyncPoint is added to help reproducing the race condition.
Test Plan: TransactionLogIteratorRace; make check
Reviewers: dhruba, ljin
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D17121
Summary: As we know, logging can be slow, or even hang for some file systems. Move one more logging out of DB mutex.
Test Plan: make all check
Reviewers: haobo, igor, ljin
Reviewed By: igor
CC: yhchiang, nkg-, leveldb
Differential Revision: https://reviews.facebook.net/D17427
Summary: The previous change D15087 changed existing compaction filter, which makes the commonly used class not backward compatible. Revert the older interface. Use a new interface for V2 instead.
Test Plan: make all check
Reviewers: haobo, yhchiang, igor
CC: danguo, dhruba, ljin, igor, leveldb
Differential Revision: https://reviews.facebook.net/D17223
Summary: Talked to <insert internal project name> folks and they found it really scary that they won't be able to roll back once they upgrade to 2.8. We should fix this.
Test Plan: make check
Reviewers: haobo, ljin
Reviewed By: ljin
CC: leveldb
Differential Revision: https://reviews.facebook.net/D17343
Summary:
NewFixedPrefixTransform is leaked in default options. Broken by b47812fba6
Also included in the diff some code cleanup
Test Plan:
valgrind env_test
also make check
Reviewers: haobo, danguo, yhchiang
Reviewed By: danguo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D17211
Summary:
AssertHeld() was a no-op before. Now it does things.
Also, this change caught a bad bug in SuperVersion::Init(). The method is calling db->mutex.AssertHeld(), but db variable is not initialized yet! I also fixed that issue.
Test Plan: make check
Reviewers: dhruba, haobo, ljin, sdong, yhchiang
Reviewed By: haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D17193
Summary:
Currently if client uses kNULLString as the prefix, it will confuse
compaction filter v2. This diff added a bool to indicate if the prefix
has been intialized. I also added a unit test to cover this case and
make sure the new code path is hit.
Test Plan: db_test
Reviewers: igor, haobo
Reviewed By: igor
CC: leveldb
Differential Revision: https://reviews.facebook.net/D17151
Summary:
This diff adds a new CompactionFilterV2 API that roll up the
decisions of kv pairs during compactions. These kv pairs must share the
same key prefix. They are buffered inside the db.
typedef std::vector<Slice> SliceVector;
virtual std::vector<bool> Filter(int level,
const SliceVector& keys,
const SliceVector& existing_values,
std::vector<std::string>* new_values,
std::vector<bool>* values_changed
) const = 0;
Application can override the Filter() function to operate
on the buffered kv pairs. More details in the inline documentation.
Test Plan:
make check. Added unit tests to make sure Keep, Delete,
Change all works.
Reviewers: haobo
CCs: leveldb
Differential Revision: https://reviews.facebook.net/D15087
Summary:
* PartialMerge api now takes a list of operands instead of two operands.
* Add min_pertial_merge_operands to Options, indicating the minimum
number of operands to trigger partial merge.
* This diff is based on Schalk's previous diff (D14601), but it also
includes necessary changes such as updating the pure C api for
partial merge.
Test Plan:
* make check all
* develop tests for cases where partial merge takes more than two
operands.
TODOs (from Schalk):
* Add test with min_partial_merge_operands > 2.
* Perform benchmarks to measure the performance improvements (can probably
use results of task #2837810.)
* Add description of problem to doc/index.html.
* Change wiki pages to reflect the interface changes.
Reviewers: haobo, igor, vamsi
Reviewed By: haobo
CC: leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D16815
Summary:
Everytime a client opens a DB, we do a sanity check that:
* checks the existance of all the necessary files
* verifies that file sizes are correct
Some of the code was stolen from https://reviews.facebook.net/D16935
Test Plan: added a unit test
Reviewers: dhruba, haobo, sdong
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D17097
Summary:
Whenever we get an IOError from GetImpl() or NewIterator(), we should immediatelly mark the DB read-only. The same check already exists in Write() and Compaction().
This should help with clients that are somehow missing a file.
Test Plan: make check
Reviewers: dhruba, haobo, sdong, ljin
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D17061
Summary: Add a property to calculate number of background errors encountered to help users build their monitoring
Test Plan: Add a unit test. make all check
Reviewers: haobo, igor, dhruba
Reviewed By: igor
CC: ljin, nkg-, yhchiang, leveldb
Differential Revision: https://reviews.facebook.net/D16959
Summary: To partly address the request @nkg- raised, add three easy-to-add properties to compactions and flushes.
Test Plan: run unit tests and add a new unit test to cover new properties.
Reviewers: haobo, dhruba
Reviewed By: dhruba
CC: nkg-, leveldb
Differential Revision: https://reviews.facebook.net/D13677
Summary:
There is a chance that an old MANIFEST is corrupted in 2.7 but just not noticed.
This check would fail them. Change it to log instead of returning a
Corruption status.
Test Plan: make
Reviewers: haobo, igor
Reviewed By: igor
CC: leveldb
Differential Revision: https://reviews.facebook.net/D16923
Summary:
Based on my recent findings (posted in our internal group), if we use fallocate without KEEP_SIZE flag, we get superior performance of fdatasync() in append-only workloads.
This diff provides an option for user to not use KEEP_SIZE flag, thus optimizing his sync performance by up to 2x-3x.
At one point we also just called posix_fallocate instead of fallocate, which isn't very fast: http://code.woboq.org/userspace/glibc/sysdeps/posix/posix_fallocate.c.html (tl;dr it manually writes out zero bytes to allocate storage). This diff also fixes that, by first calling fallocate and then posix_fallocate if fallocate is not supported.
Test Plan: make check
Reviewers: dhruba, sdong, haobo, ljin
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D16761
Summary:
When the manifest is getting rolled the following happens:
1) manifest_file_number_ is assigned to a new manifest number (even though the old one is still current)
2) mutex is unlocked
3) SetCurrentFile() creates temporary file manifest_file_number_.dbtmp
4) SetCurrentFile() renames manifest_file_number_.dbtmp to CURRENT
5) mutex is locked
If FindObsoleteFiles happens between (3) and (4) it will:
1) Delete manifest_file_number_.dbtmp (because it's not in pending_outputs_)
2) Delete old manifest (because the manifest_file_number_ already points to a new one)
I introduce the concept of prev_manifest_file_number_ that will avoid the race condition.
However, we should discuss the future of MANIFEST file rolling. We found some race conditions with it last week and who knows how many more are there. Nobody is using it in production because we don't trust the implementation. Should we even support it?
Test Plan: make check
Reviewers: ljin, dhruba, haobo, sdong
Reviewed By: haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D16929
Summary: Client doesn't need to know anything about ColumnFamily ID. By making WriteBatch take ColumnFamilyHandle as a parameter, we can eliminate method GetID() from ColumnFamilyHandle
Test Plan: column_family_test
Reviewers: haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D16887
Summary:
With D16767, there is a case compaction tasks are scheduled infinitely:
(1) no flush thread is configured and more than 1 compaction threads
(2) a flush is going on by one compaction hread
(3) the state of SST files is in the state that versions_->current()->NeedsCompaction() will generate a false positive (return true actually there is no work to be done)
In that case, a infinite loop will be formed.
This patch would fix it.
Test Plan: make all check
Reviewers: haobo, igor, ljin
Reviewed By: igor
CC: dhruba, yhchiang, leveldb
Differential Revision: https://reviews.facebook.net/D16863
Summary:
This is is based on https://reviews.facebook.net/D15027. It's not finished but I would like to give a prototype to avoid arena over-allocation while making better use of the already allocated memory blocks.
Instead of check approximate memtable size, we will take a deeper look at the arena, which incorporate essential idea that @sdong suggests: flush when arena has allocated its last and the last is "almost full"
Test Plan: N/A
Reviewers: haobo, sdong
Reviewed By: sdong
CC: leveldb, sdong
Differential Revision: https://reviews.facebook.net/D15051
Summary:
I'm cleaning up some code preparing for the big diff review tomorrow. This is the first part of the cleanup.
Changes are mostly cosmetic. The goal is to decrease amount of code difference between columnfamilies and master branch.
This diff also fixes race condition when dropping column family.
Test Plan: Ran db_stress with variety of parameters
Reviewers: dhruba, haobo
Differential Revision: https://reviews.facebook.net/D16833
Summary:
@igor pointed out that there is a potential data race because of the way we use the newly introduced LogBuffer. After "bg_compaction_scheduled_--" or "bg_flush_scheduled_--", they can both become 0. As soon as the lock is released after that, DBImpl's deconstructor can go ahead and deconstruct all the states inside DB, including the info_log object hold in a shared pointer of the options object it keeps. At that point it is not safe anymore to continue using the info logger to write the delayed logs.
With the patch, lock is released temporarily for log buffer to be flushed before "bg_compaction_scheduled_--" or "bg_flush_scheduled_--". In order to make sure we don't miss any pending flush or compaction, a new flag bg_schedule_needed_ is added, which is set to be true if there is a pending flush or compaction but not scheduled because of the max thread limit. If the flag is set to be true, the scheduling function will be called before compaction or flush thread finishes.
Thanks @igor for this finding!
Test Plan: make all check
Reviewers: haobo, igor
Reviewed By: haobo
CC: dhruba, ljin, yhchiang, igor, leveldb
Differential Revision: https://reviews.facebook.net/D16767
Summary:
I had this diff for a while to test column families implementation. Last night, I ran it sucessfully for 10 hours with the command:
time ./db_stress --threads=30 --ops_per_thread=200000000 --max_key=5000 --column_families=20 --clear_column_family_one_in=3000000 --verify_before_write=1 --reopen=50 --max_background_compactions=10 --max_background_flushes=10 --db=/tmp/db_stress
It is ready to be committed :)
Test Plan: Ran it for 10 hours
Reviewers: dhruba, haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D16797
Summary: To temp fix the log buffer flushing. Flush the buffer inside the lock. Clean the trunk before we find an eventual fix.
Test Plan: make all check
Reviewers: haobo, igor
Reviewed By: igor
CC: ljin, leveldb, yhchiang
Differential Revision: https://reviews.facebook.net/D16791
Summary: Having code after SignalAll has already caused 2 bugs. Let's make sure this doesn't happen again.
Test Plan: no test
Reviewers: sdong, dhruba, haobo
Reviewed By: haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D16785
Summary: as title. also made info log output of file deletion a bit more descriptive.
Test Plan: make check; db_bench and look at LOG output
Reviewers: igor
Reviewed By: igor
CC: leveldb
Differential Revision: https://reviews.facebook.net/D16731
Summary:
(1) Fix SanitizeOptions() to also check HashLinkList. The current
dynamic case just happens to work because the 2 classes have the same
layout.
(2) Do not delete SliceTransform object in HashSkipListFactory and
HashLinkListFactory destructor. Reason: SanitizeOptions() enforces
prefix_extractor and SliceTransform to be the same object when
Hash**Factory is used. This makes the behavior strange: when
Hash**Factory is used, prefix_extractor will be released by RocksDB. If
other memtable factory is used, prefix_extractor should be released by
user.
Test Plan: db_bench && make asan_check
Reviewers: haobo, igor, sdong
Reviewed By: igor
CC: leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D16587
Summary: KSVObsolete is no longer nullptr and needs to be checked explicitly. Also did some minor code cleanup and added a stat counter to track superversion cleanups incurred in the foreground.
Test Plan: make check
Reviewers: ljin
Reviewed By: ljin
CC: leveldb
Differential Revision: https://reviews.facebook.net/D16701
Summary: Moved LogBuffer class to an internal header. Removed some unneccesary indirection. Enabled log buffer for BackgroundCallFlush. Forced log buffer flush right after Unlock to improve time ordering of info log.
Test Plan: make check; db_bench compare LOG output
Reviewers: sdong
Reviewed By: sdong
CC: leveldb, igor
Differential Revision: https://reviews.facebook.net/D16707
Summary: Adding the last missing function -- NewIterators(). Pretty simple implementation
Test Plan: added a unit test
Reviewers: dhruba, haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D16689
Summary:
Add a check at the end of GetImpl to release SuperVersion if it becomes
obsolete. Also do Scrape() inside InstallSuperVersion so it happens more
frequent.
Test Plan:
make all check
running asan_check now
Reviewers: igor, haobo, sdong, dhruba
Reviewed By: haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D16641
Summary: Now while the background thread is picking compactions, it writes out multiple info_logs, especially for universal compaction, which introduces a chance of waiting log writing in mutex, which is bad. To remove this risk, write all those info logs to a buffer and flush it after releasing the mutex.
Test Plan:
make all check
check the log lines while running some tests that trigger compactions.
Reviewers: haobo, igor, dhruba
Reviewed By: dhruba
CC: i.am.jin.lei, dhruba, yhchiang, leveldb, nkg-
Differential Revision: https://reviews.facebook.net/D16515
Summary:
Two changes:
1. DeletionState is only constructed when cleaning up is needed
2. Fix the bug of deletion state construction bug. A change was made in a previous patch: https://reviews.facebook.net/rROCKSDB774ed89c2405ee058086b099cbc8b29e243739cc#71a34e2e However, it somehow got lost when merging
Test Plan: make all check
Reviewers: kailiu, haobo, igor
Reviewed By: igor
CC: igor, dhruba, i.am.jin.lei, yhchiang, leveldb
Differential Revision: https://reviews.facebook.net/D16233
Summary: When column family is dropped, we want to delete all WALs that refer to it. To do that, we need to make them obsolete by flushing all the memtables
Test Plan: column_family_test
Reviewers: dhruba, haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D16557
Summary: Added a function DeleteSuperVersion that can be called in DBImpl destructor before PurgingObsoleteFiles. That way, PurgeObsoleteFiles will be able to delete all files held by alive super versions.
Test Plan: column_family_test with valgrind
Reviewers: dhruba, haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D16545
Summary:
This diff fixes two bugs:
* Increase sequence number even if WriteBatch fails. This is important because WriteBatches in WAL logs have implictly increasing sequence number, even if one update in a write batch fails. This caused some writes to get lost in my CF stress testing
* Tolerate 'invalid column family' errors on recovery. When a column family is dropped, processing WAL logs can have some WriteBatches that still refer to the dropped column family. In recovery environment, we want to ignore those errors. In client's Write() code path, however, we want to return the failure to the client if he's trying to add data to invalid column family.
Test Plan: db_stress's verification works now
Reviewers: dhruba, haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D16533
Summary:
I though I might get away with as little changes to LogAndApply() as possible. It turns out this is not the case.
This diff introduces different behavior of LogAndApply() for three cases:
1. column family add
2. column family drop
3. no-column family manipulation
(1) and (2) don't support group commit yet.
There were a lot of problems with old version od LogAndApply, detected by db_stress. The biggest was non-atomicity of manifest writes and metadata changes (i.e. if column family add is in manifest, it also has to be in in-memory data structure).
Test Plan: db_stress
Reviewers: dhruba, haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D16491
Summary:
This diff does two things:
(1) Log::Reader does not report a corruption when the last record in a log or manifest file is truncated (meaning that log writer died in the middle of the write). Inherited the code from LevelDB: https://code.google.com/p/leveldb/source/detail?r=269fc6ca9416129248db5ca57050cd5d39d177c8#
(2) Turn off mmap writes for all writes to log and manifest files
(2) is necessary because if we use mmap writes, the last record is not truncated, but is actually filled with zeros, making checksum fail. It is hard to recover from checksum failing.
Test Plan:
Added unit tests from LevelDB
Actually recovered a "corrupted" MANIFEST file.
Reviewers: dhruba, haobo
Reviewed By: haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D16119
Summary:
Add an optional input parameter ReadOptions to DB::GetUpdateSince(),
which allows the verification of checksums to be disabled by setting
ReadOptions::verify_checksums to false.
Test Plan: Tests are done off-line and will not be included in the regular unit test.
Reviewers: igor
Reviewed By: igor
CC: leveldb, xjin, dhruba
Differential Revision: https://reviews.facebook.net/D16305
Summary:
This fixes few bugs with CreateColumnFamily
* We first have to LogAndApply and then call VersionSet::CreateColumnFamily. Otherwise, WriteSnapshot might be invoked, writing out column family add inside of LogAndApply, even though it's not really committed
* Fix LogAndApplyHelper() to not apply log number to column_family_data, which is in case of column family add, just a dummy (default) column family
* Create SuperVerion when creating column family
Test Plan: column_family_test
Reviewers: dhruba, haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D16443
Summary:
Previously, we first wrote to the manifest and then created internal data structure.
Now, we first create internal data structure. That way, we can write out internal comparator to the manifest
Test Plan: column_family_test
Reviewers: dhruba, haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D16425
Summary: as title
Test Plan:
asan_check
will post results later
Reviewers: haobo, igor, dhruba, sdong
Reviewed By: haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D16257
Summary:
Two new column family tests:
* DifferentMergeOperators -- three column families, one without merge operator, one with add operator and one with append operator. verify that operations work as expected.
* DifferentCompactionStyles -- three column families, two with level compactions and one with universal compaction. trigger the compactions and verify they work as expected.
Test Plan: nope
Reviewers: dhruba, haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D16377
Summary: New unit tests for column families
Test Plan: this is a test
Reviewers: dhruba, haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D16359
Summary: Big CF diff uncovered some lint errors. This diff fixes some of them. Not much to see here
Test Plan: make check
Reviewers: dhruba, haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D16347
Summary:
* Add ColumnFamilyHandle::GetID() function. Client needs to know column family's ID to be able to construct WriteBatch
* Handle WriteBatch::Handler failure gracefully. Since WriteBatch is not a very smart function (it takes raw CF id), client can add data to WriteBatch for column family that doesn't exist. In that case, we need to gracefully return failure status from DB::Write(). To do that, I added a return Status to WriteBatch functions PutCF, DeleteCF and MergeCF.
Test Plan: Added test to column_family_test
Reviewers: dhruba, haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D16323
Summary:
* Added unit test that verifies that obsolete files are deleted.
* Advance log number for empty column family when cutting log file.
* MinLogNumber() bug fix! (caught by the new unit test)
Test Plan: unit test
Reviewers: dhruba, haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D16311
Summary:
This will also help with avoiding the deadlock. If a flush failed and we're waiting for a memtable to be flushed, we should schedule a new flush and hope a new one succeedes.
If paranoid_checks = false, Wait() will still hang on ENOSPC, but at least it will automatically continue when the space frees up. Current behavior both hangs and deadlocks.
Also, I renamed some 'compaction' to 'flush'. 'compaction' was leveldb way of saying things.
Test Plan: make check
Reviewers: dhruba, haobo, ljin
Reviewed By: haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D16281
Summary: DBImpl now keeps a list of alive_log_files_. On every FindObsoleteFiles, it deletes all alive log files that are smaller than versions_->MinLogNumber()
Test Plan:
make check passes
no specific unit tests yet, will add
Reviewers: dhruba, haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D16293
Summary:
More info here: https://github.com/facebook/rocksdb/issues/89
If flush fails because of ENOSPC, we have a deadlock problem. This is a quick fix that will continue the normal operation when user deletes the file and frees up the space on the device.
We need to address the issue more broadly with bg_error_ cleanup.
Test Plan: make check
Reviewers: dhruba, haobo, ljin
Reviewed By: ljin
CC: leveldb
Differential Revision: https://reviews.facebook.net/D16275
Summary: Currently, the first transaction log file ignore bytes_per_sync and other storage-related options. It is not consistent. Fix it.
Test Plan: make all check. See the options set in GDB.
Reviewers: haobo, kailiu
Reviewed By: haobo
CC: igor, ljin, yhchiang, leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D16215
Summary: Adapt table properties to column family world
Test Plan: make check
Reviewers: kailiu
CC: leveldb
Differential Revision: https://reviews.facebook.net/D16161
Summary:
This is a huge diff and it was hectic, but the idea is actually quite simple. Every operation (Put, Get, etc.) done on default column family in DBTest is now forwarded to non-default ("pikachu"). The good news is that we had zero test failures! Column families look stable so far.
One interesting test that I adapted for column families is MultiThreadedTest. I replaced every Put() with a WriteBatch writing to all column families concurrently. Every Put in the write batch contains unique_id. Instead of Get() I do a multiget across all column families with the same key. If atomicity holds, I expect to see the same unique_id in all column families.
Test Plan: This is a test!
Reviewers: dhruba, haobo, kailiu, sdong
CC: leveldb
Differential Revision: https://reviews.facebook.net/D16149
Summary: Provide a public API for users to access the table properties for each SSTable.
Test Plan: Added a unit tests to test the function correctness under differnet conditions.
Reviewers: haobo, dhruba, sdong
Reviewed By: haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D16083
Summary:
1. Add some more implementation-aware tests for PlainTable
2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks.
3. Explicitly support total order (only use binary search)
4. some code cleaning up.
Test Plan: make all check
Reviewers: haobo, kailiu
Reviewed By: haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D16023
Summary:
The change to the public behavior:
* When opening a DB or creating new column family client gets a ColumnFamilyHandle.
* As long as column family handle is alive, client can do whatever he wants with it, even drop it
* Dropped column family can still be read from (using the column family handle)
* Added a new call CloseColumnFamily(). Client has to close all column families that he has opened before deleting the DB
* As soon as column family is closed, any calls to DB using that column family handle will fail (also any outstanding calls)
Internally:
* Ref-counting ColumnFamilyData
* New thread-safety for ColumnFamilySet
* Dropped column families are now completely dropped and their memory cleaned-up
Test Plan: added some tests to column_family_test
Reviewers: dhruba, haobo, kailiu, sdong
CC: leveldb
Differential Revision: https://reviews.facebook.net/D16101
Summary: Added a bit more information to compaction context, requested by internal team at FB.
Test Plan: Modified CompactionFilter test to make sure is_manual_compaction is properly set.
Reviewers: haobo
Reviewed By: haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D16095
Summary: Clean up IOErrors so that it only indicates errors talking to device.
Test Plan: make all check
Reviewers: igor, haobo, dhruba, emayanke
Reviewed By: igor
CC: leveldb
Differential Revision: https://reviews.facebook.net/D15831
Summary: This covers existing table files before DB open happens and avoids contention on table cache
Test Plan: db_test
Reviewers: haobo, sdong, igor, dhruba
Reviewed By: haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D16089
Summary: By discussion with @dhruba, overloading Open makes more sense
Test Plan: compiles!
Reviewers: dhruba
CC: leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D16017
Summary: When we open a DB, we should dump only DBOptions and then when we create a new column family, we dump ColumnFamilyOptions for each one.
Test Plan: make check, confirm contents of the LOG
Reviewers: dhruba, haobo, sdong, kailiu
CC: leveldb
Differential Revision: https://reviews.facebook.net/D16011
Summary:
This is not column-family related diff. It is in columnfamily branch because the change is significant and we want to push it with next major release (3.0).
It removes the leveldb notion of one thread pool and expands it to two thread pools by default (HIGH and LOW). Flush process is removed from compaction process and all flush threads are executed on HIGH thread pool, since we don't want long-running compactions to influence flush latency.
Test Plan: make check
Reviewers: dhruba, haobo, kailiu, sdong
CC: leveldb
Differential Revision: https://reviews.facebook.net/D15987
Summary: Revised thread-safety guarantees and implemented a way to spinlock the object.
Test Plan: make check
Reviewers: dhruba, haobo, sdong, kailiu
CC: leveldb
Differential Revision: https://reviews.facebook.net/D15975
Summary:
WriteBatch can have multiple column families in one batch. Every column family has different options. So we have to add a way for write batch to get options for an arbitrary column family.
This required a bit more acrobatics since lots of interfaces had to be changed.
Test Plan: make check
Reviewers: dhruba, haobo, sdong, kailiu
CC: leveldb
Differential Revision: https://reviews.facebook.net/D15957
Summary: Replaced most of occurrences of Options with more specific DBOptions. This brings us very close to supporting different configuration options for each column family.
Test Plan: make check
Reviewers: dhruba, haobo, kailiu, sdong
CC: leveldb
Differential Revision: https://reviews.facebook.net/D15933
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
Summary: InternalStats is a messy thing, keeping both DB data and column family data. However, it's better off living in ColumnFamilyData than in DBImpl. For now, at least.
Test Plan: make check
Reviewers: dhruba, kailiu, haobo, sdong
CC: leveldb
Differential Revision: https://reviews.facebook.net/D15879
Summary:
There are three SanitizeOption-s now : one for DBOptions, one for ColumnFamilyOptions and one for Options (which just calls the other two)
I have also reshuffled some options -- table_cache options and info_log should live in DBOptions, for example.
Test Plan: make check doesn't complain
Reviewers: dhruba, haobo, kailiu, sdong
CC: leveldb
Differential Revision: https://reviews.facebook.net/D15873
Summary: user_comparator() is a Column Family property, not DBImpl
Test Plan: make check
Reviewers: dhruba, haobo, kailiu, sdong
CC: leveldb
Differential Revision: https://reviews.facebook.net/D15855
Summary: Support for different column families in Iterator and MultiGet code path.
Test Plan: make check
Reviewers: dhruba, haobo, kailiu, sdong
CC: leveldb
Differential Revision: https://reviews.facebook.net/D15849
Summary:
Use super_version insider NewIterator to avoid Ref() each component
separately under mutex
The new added bench shows NewIterator QPS increases from 515K to 719K
No meaningful improvement for multiget I guess due to its relatively small
cost comparing to 90 keys fetch in the test.
Test Plan: unit test and db_bench
Reviewers: igor, sdong
Reviewed By: igor
CC: leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D15609
Summary: In PlainTable, use one single byte to represent 8 bytes of internal bytes, if seqID = 0 and it is value type (which should be common for bottom most files). It is to save 7 bytes for uncompressed cases.
Test Plan: make all check
Reviewers: haobo, dhruba, kailiu
Reviewed By: haobo
CC: igor, leveldb
Differential Revision: https://reviews.facebook.net/D15489
Summary: This diff enables non-default column families to get compacted both automatically and also by calling CompactRange()
Test Plan: make check
Reviewers: dhruba, haobo, kailiu, sdong
CC: leveldb
Differential Revision: https://reviews.facebook.net/D15813
Summary: Compaction picker and internal key comparator are different for each column family (not global), so they should live in ColumnFamilyData
Test Plan: make check
Reviewers: dhruba, haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D15801
Summary:
VersionSet::next_file_number_ is always assumed to be strictly greater than VersionSet::log_number_. In our new recovery code, we artificially set log_number_ to be (log_number + 1), so that once we flush, we don't recover from the same log file again (this is important because of merge operator non-idempotence)
When we set VersionSet::log_number_ to (log_number + 1), we also have to mark that file number used, such that next_file_number_ is increased to a legal level. Otherwise, VersionSet might assert.
This has not be a problem so far because here's what happens:
1. assume next_file_number is 5, we're recovering log_number 10
2. in DBImpl::Recover() we call MarkFileNumberUsed with 10. This will set VersionSet::next_file_number_ to 11.
3. If there are some updates, we will call WriteTable0ForRecovery(), which will use file number 11 as a new table file and advance VersionSet::next_file_number_ to 12.
4. When we LogAndApply() with log_number 11, assertion is true: assert(11 <= 12);
However, this was a lucky occurrence. Even though this diff doesn't cause a bug, I think the issue is important to fix.
Test Plan: In column families I have different recovery logic and this code path asserted. When adding MarkFileNumberUsed(log_number + 1) assert is gone.
Reviewers: dhruba, kailiu
Reviewed By: kailiu
CC: leveldb
Differential Revision: https://reviews.facebook.net/D15783
Summary: Removed default_cfd_ from all flush code paths. This means we can now flush memtables from arbitrary column families!
Test Plan: Added a new unit test
Reviewers: dhruba, haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D15789
Summary: I didn't figure out the reason why the feature of zeroing out earlier sequence ID is disabled in universal compaction. I do see bottommost_level is set correctly. It should simply work if we remove the constraint of universal compaction.
Test Plan: make all check
Reviewers: haobo, dhruba, kailiu, igor
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D15423
Summary:
Sometimes we iterate through column families, and unlock the mutex in the body of the iteration. While mutex is unlocked, some column family might be created or dropped. We need to be able to continue iterating through column families even though our current column family got dropped.
This diff implements circular linked lists that connect all column families. It then uses the link list to enable iterating through linked lists. Even if the column family is dropped, its next_ pointer still can be used to advance to another alive column family.
Test Plan: make check
Reviewers: dhruba, haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D15603
Summary: Making room for write will be the hardest part of the column family implementation. For now, I just iterate through all column families and run MakeRoomForWrite() for every one.
Test Plan: make check does not complain
Reviewers: dhruba, haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D15597
Summary:
In DBImpl we keep track of some statistics internally and expose them via GetProperty(). This diff encapsulates all the internal statistics into a class InternalStatisics. Most of it is copy/paste.
Apart from cleaning up db_impl.cc, this diff is also necessary for Column families, since every column family should have its own CompactionStats, MakeRoomForWrite-stall stats, etc. It's much easier to keep track of it in every column family if it's nicely encapsulated in its own class.
Test Plan: make check
Reviewers: dhruba, kailiu, haobo, sdong, emayanke
Reviewed By: haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D15273
Summary: as title
Test Plan: unit test
Reviewers: haobo, igor, sdong, kailiu, dhruba
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D15435
Summary: ColumnFamilyData grew a lot, there's much more data that it holds now. It makes more sense to encapsulate it better by making it a class.
Test Plan: make check
Reviewers: dhruba, haobo, kailiu, sdong
CC: leveldb
Differential Revision: https://reviews.facebook.net/D15579
Summary: When we drop the column family, we want to delete all the files from that column family.
Test Plan: make check
Reviewers: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D15561
Summary: This one is big. It adds ability to write to and read from different column families (see the unit test). It also supports recovery of different column families from log, which was the hardest part to reason about. We need to make sure to never delete the log file which has unflushed data from any column family. To support that, I added another concept, which is versions_->MinLogNumber()
Test Plan: Added a unit test in column_family_test
Reviewers: dhruba, haobo, sdong, kailiu
CC: leveldb
Differential Revision: https://reviews.facebook.net/D15537
Summary: This removes the default implementation of LogAndApply that applied the changed to the default column family by default. It is mostly simple reformatting.
Test Plan: make check
Reviewers: dhruba, kailiu
CC: leveldb
Differential Revision: https://reviews.facebook.net/D15465
Summary: All memtables and immutable memtables are moved from DBImpl to ColumnFamilyData. For now, they are all referenced from default column family in DBImpl. It shouldn't be hard to get them from custom column family.
Test Plan: make check
Reviewers: dhruba, kailiu, sdong
CC: leveldb
Differential Revision: https://reviews.facebook.net/D15459
Summary:
@dhruba, I'm not sure where we need to sync the directory. I implemented the function in Env() and added the dir sync just after we close the newly created file in the builder.
Should I also add FsyncDir() to new files that get created by a compaction?
Test Plan: Confirmed that FsyncDir is returning Status::OK()
Reviewers: dhruba, haobo
Reviewed By: dhruba
CC: leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D14751
Summary: There is no reason to have functions NeedCompaction(), MaxCompactionScore() and MaxCompactionScoreLevel() in VersionSet, since they don't access any data in VersionSet.
Test Plan: make check
Reviewers: kailiu, haobo, sdong
Reviewed By: kailiu
CC: leveldb
Differential Revision: https://reviews.facebook.net/D15333
Summary:
MemTableListVersion is to MemTableList what Version is to VersionSet. I took almost the same ideas to develop MemTableListVersion. The reason is to have copying std::list done in background, while flushing, rather than in foreground (MultiGet() and NewIterator()) under a mutex! Also, whenever we copied MemTableList, we copied also some MemTableList metadata (flush_requested_, commit_in_progress_, etc.), which was wasteful.
This diff avoids std::list copy under a mutex in both MultiGet() and NewIterator(). I created a small database with some number of immutable memtables, and creating 100.000 iterators in a single-thread (!) decreased from {188739, 215703, 198028} to {154352, 164035, 159817}. A lot of the savings come from code under a mutex, so we should see much higher savings with multiple threads. Creating new iterator is very important to LogDevice team.
I also think this diff will make SuperVersion obsolete for performance reasons. I will try it in the next diff. SuperVersion gave us huge savings on Get() code path, but I think that most of the savings came from copying MemTableList under a mutex. If we had MemTableListVersion, we would never need to copy the entire object (like we still do in NewIterator() and MultiGet())
Test Plan: `make check` works. I will also do `make valgrind_check` before commit
Reviewers: dhruba, haobo, kailiu, sdong, emayanke, tnovak
Reviewed By: kailiu
CC: leveldb
Differential Revision: https://reviews.facebook.net/D15255
Summary: as title
Test Plan:
make all check
What else tests shall I cover?
Reviewers: igor, haobo
CC:
Differential Revision: https://reviews.facebook.net/D15339
Summary:
This diff implements a special type of iterator that doesn't create a snapshot
(can be used to read newly inserted data) and is optimized for doing sequential
reads.
TailingIterator uses current superversion number to determine whether to
invalidate its internal iterators. If the version hasn't changed, it can often
avoid doing expensive seeks over immutable structures (sst files and immutable
memtables).
Test Plan:
* new unit tests
* running LD with this patch
Reviewers: igor, dhruba, haobo, sdong, kailiu
Reviewed By: sdong
CC: leveldb, lovro, march
Differential Revision: https://reviews.facebook.net/D15285
Summary:
I created a separate class ColumnFamilySet to keep track of column families. Before we did this in VersionSet and I believe this approach is cleaner.
Let me know if you have any comments. I will commit tomorrow.
Test Plan: make check
Reviewers: dhruba, haobo, kailiu, sdong
CC: leveldb
Differential Revision: https://reviews.facebook.net/D15357
Summary:
This diff does two things:
* Rethinks how we call Recover() with read_only option. Before, we call it with pointer to memtable where we'd like to apply those changes to. This memtable is set in db_impl_readonly.cc and it's actually DBImpl::mem_. Why don't we just apply updates to mem_ right away? It seems more intuitive.
* Changes when we apply updates to manifest. Before, the process is to recover all the logs, flush it to sst files and then do one giant commit that atomically adds all recovered sst files and sets the next log number. This works good enough, but causes some small troubles for my column family approach, since I can't have one VersionEdit apply to more than single column family[1]. The change here is to commit the files recovered from logs right away. Here is the state of the world before the change:
1. Recover log 5, add new sst files to edit
2. Recover log 7, add new sst files to edit
3. Recover log 8, add new sst files to edit
4. Commit all added sst files to manifest and mark log files 5, 7 and 8 as recoverd (via SetLogNumber(9) function)
After the change, we'll do:
1. Recover log 5, commit the new sst files and set log 5 as recovered
2. Recover log 7, commit the new sst files and set log 7 as recovered
3. Recover log 8, commit the new sst files and set log 8 as recovered
The added (small) benefit is that if we fail after (2), the new recovery will only have to recover log 8. In previous case, we'll have to restart the recovery from the beginning. The bigger benefit will be to enable easier integration of multiple column families in Recovery code path.
[1] I'm happy to dicuss this decison, but I believe this is the cleanest way to go. It also makes backward compatibility much easier. We don't have a requirement of adding multiple column families atomically.
Test Plan: make check
Reviewers: dhruba, haobo, kailiu, sdong
Reviewed By: kailiu
CC: leveldb
Differential Revision: https://reviews.facebook.net/D15237
Summary:
This moves the use of versions_ to before the mutex is unlocked
to avoid a possible race.
Task ID: #
Blame Rev:
Test Plan:
make check
Revert Plan:
Database Impact:
Memcache Impact:
Other Notes:
EImportant:
- begin *PUBLIC* platform impact section -
Bugzilla: #
- end platform impact -
Reviewers: haobo, dhruba
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D15279
Summary: I'm separating code-cleanup part of https://reviews.facebook.net/D14517. This will make D14517 easier to understand and this diff easier to review.
Test Plan: make check
Reviewers: haobo, kailiu, sdong, dhruba, tnovak
Reviewed By: tnovak
CC: leveldb
Differential Revision: https://reviews.facebook.net/D15099
Summary:
This had a few bugs.
1) bottom and top were reversed. top is for the max value but the callers were passing the max
value to bottom. The result is that the max sleep is used when n >= bottom.
2) one of the callers passed values with type double and these values are frequently between
1.0 and 2.0 so rounding will do some bad things
3) sometimes the function returned 0 when there should be a stall
With this change and one other diff (out for review soon) there are slightly fewer stalls on one workload.
With the fix.
Stalls(secs): 160.166 level0_slowdown, 0.000 level0_numfiles, 0.000 memtable_compaction, 58.495 leveln_slowdown
Stalls(count): 910261 level0_slowdown, 0 level0_numfiles, 0 memtable_compaction, 54526 leveln_slowdown
Without the fix.
Stalls(secs): 172.227 level0_slowdown, 0.000 level0_numfiles, 0.000 memtable_compaction, 56.538 leveln_slowdown
Stalls(count): 160831 level0_slowdown, 0 level0_numfiles, 0 memtable_compaction, 52845 leveln_slowdown
Task ID: #
Blame Rev:
Test Plan:
run db_bench for --benchmarks=overwrite with IO-bound database
Revert Plan:
Database Impact:
Memcache Impact:
Other Notes:
EImportant:
- begin *PUBLIC* platform impact section -
Bugzilla: #
- end platform impact -
Reviewers: haobo
Reviewed By: haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D15243
Summary:
There were some functions in VersionSet that had no reason to be there instead of Version. Moving them to Version will make column families implementation easier.
The functions moved are:
* NumLevelBytes
* LevelSummary
* LevelFileSummary
* MaxNextLevelOverlappingBytes
* AddLiveFiles (previously AddLiveFilesCurrentVersion())
* NeedSlowdownForNumLevel0Files
The diff continues on (and depends on) D15171
Test Plan: make check
Reviewers: dhruba, haobo, kailiu, sdong, emayanke
Reviewed By: sdong
CC: leveldb
Differential Revision: https://reviews.facebook.net/D15183
Summary:
With column families VersionSet will not have a constant number of levels (each CF can have different options), so we'll need to eliminate call to VersionSet::NumberLevels()
This diff decreases number of callsites, but we're not there yet. It associates number of levels with Version (each version is associated with single CF) instead of VersionSet.
I have also slightly changed how VersionSet keeps track of manifest size.
This diff also modifies constructor of Compaction such that it takes input_version and automatically Ref()s it. Before this was done outside of constructor.
In next diffs I will continue to decrease number of callsites of VersionSet::NumberLevels() and also references to current_
Test Plan: make check
Reviewers: haobo, dhruba, kailiu, sdong
Reviewed By: sdong
Differential Revision: https://reviews.facebook.net/D15171
Summary: To reduce mutex contention caused by DBImpl.NewInternalIterator(), in this function, move all the iteration creation works out of mutex, only leaving object ref and get.
Test Plan:
make all check
will run db_stress for a while too to make sure no problem.
Reviewers: haobo, dhruba, kailiu
Reviewed By: haobo
CC: igor, leveldb
Differential Revision: https://reviews.facebook.net/D14589
Conflicts:
db/db_impl.cc
Summary:
When doing CompactRange(), we should first flush the memtable and then calculate max_level_with_files. Also, we want to compact all the levels that have files, including level `max_level_with_files`.
This patch fixed the unit test.
Test Plan: Added a failing unit test and a fix, so it's not failing anymore.
Reviewers: dhruba, haobo, sdong
Reviewed By: haobo
CC: leveldb, xjin
Differential Revision: https://reviews.facebook.net/D14421
Summary:
I will submit a sequence of diffs that are preparing master branch for column families. There are a lot of implicit assumptions in the code that are making column family implementation hard. If I make the change only in column family branch, it will make merging back to master impossible.
Most of the diffs will be simple code refactorings, so I hope we can have fast turnaround time. Feel free to grab me in person to discuss any of them.
This diff removes number of level check from VersionEdit. It is used only when VersionEdit is read, not written, but has to be set when it is written. I believe it is a right thing to make VersionEdit dumb and check consistency on the caller side. This will also make it much easier to implement Column Families, since different column families can have different number of levels.
Test Plan: make check
Reviewers: dhruba, haobo, sdong, kailiu
Reviewed By: kailiu
CC: leveldb
Differential Revision: https://reviews.facebook.net/D15159
Summary: When building batch group, don't actually build a new batch since it requires heavy-weight mem copy and malloc. Only store references to the batches and build the batch group without lock held.
Test Plan:
`make check`
I am also planning to run performance tests. The workload that will benefit from this change is readwhilewriting. I will post the results once I have them.
Reviewers: dhruba, haobo, kailiu
Reviewed By: haobo
CC: leveldb, xjin
Differential Revision: https://reviews.facebook.net/D15063
Summary: We use SanitizeOptions() to set appropriate values for some options, based on other options. So we should use the sanitized options by default. Luckily it hasn't caused a bug yet, but can result in a bug in the fugture.
Test Plan: make check
Reviewers: haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D14103
Summary:
In one of CPU profiles, we see some CPU costs of string::reserve() inside Batch.Put(). This patch should be able to reduce some of the costs by allocating sufficient buffer before hand.
Since it is a trivial percentage of CPU costs, I didn't find a way to show the improvement in one of the benchmarks. I'll deploy it to same application and do the same CPU profiling to make sure those CPU costs are reduced.
Test Plan: make all check
Reviewers: haobo, kailiu, igor
Reviewed By: haobo
CC: leveldb, nkg-
Differential Revision: https://reviews.facebook.net/D15135
Summary: Currently in DBImpl::MakeRoomForWrite(), we do "versions_->NumLevelFiles(0) >= options_.level0_slowdown_writes_trigger" to check whether the writer thread needs to slow down. However, versions_->NumLevelFiles(0) is slightly more expensive than we expected. By caching the result of the comparison when installing a new version, we can avoid this function call every time.
Test Plan:
make all check
Manually trigger this behavior by applying universal compaction style and make sure inserts are made slow after there are certain number of files.
Reviewers: haobo, kailiu, igor
Reviewed By: kailiu
CC: nkg-, leveldb
Differential Revision: https://reviews.facebook.net/D15141
Summary:
In one of CPU profiles, we see some CPU costs of string::reserve() inside Batch.Put(). This patch should be able to reduce some of the costs by allocating sufficient buffer before hand.
Since it is a trivial percentage of CPU costs, I didn't find a way to show the improvement in one of the benchmarks. I'll deploy it to same application and do the same CPU profiling to make sure those CPU costs are reduced.
Test Plan: make all check
Reviewers: haobo, kailiu, igor
Reviewed By: haobo
CC: leveldb, nkg-
Differential Revision: https://reviews.facebook.net/D15135
Summary:
The biggest change here is getting rid of current_ Version and adding a column_family_data->current Version to each column family.
I have also fixed some smaller things in VersionSet that made it easier to implement Column family support.
Test Plan: make check
Reviewers: dhruba, haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D15105
Summary:
In some use cases, table readers for all live files should always be cached. In that case, there will be an opportunity to avoid the table cache look-up while Get() and NewIterator().
We define options.max_open_files = -1 to be the mode that table readers for live files will always be kept. In that mode, table readers are cached in FileMetaData (with a reference count hold in table cache). So that when executing table_cache.Get() and table_cache.newInterator(), LRU cache checking can be by-passed, to reduce latency.
Test Plan: add a test case in db_test
Reviewers: haobo, kailiu
Reviewed By: haobo
CC: dhruba, igor, leveldb
Differential Revision: https://reviews.facebook.net/D15039
Summary: Currently, even if statistics is not enabled, StopWatch only for the stats still gets the time of the day, which is wasteful. This patch adds a new option to StopWatch to disable this get in this case.
Test Plan: make all check
Reviewers: dhruba, haobo, igor
CC: leveldb
Differential Revision: https://reviews.facebook.net/D14703
Conflicts:
db/db_impl.cc
Summary: Currently, even if statistics is not enabled, StopWatch only for the stats still gets the time of the day, which is wasteful. This patch adds a new option to StopWatch to disable this get in this case.
Test Plan: make all check
Reviewers: dhruba, haobo, igor
CC: leveldb
Differential Revision: https://reviews.facebook.net/D14703
Summary:
I have added three new value types:
* kTypeColumnFamilyDeletion
* kTypeColumnFamilyValue
* kTypeColumnFamilyMerge
which include column family Varint32 before the data (value, deletion and merge). These values are used only in WAL (not in memtables yet).
This endeavour required changing some WriteBatch internals.
Test Plan: Added a unittest
Reviewers: dhruba, haobo, sdong, kailiu
CC: leveldb
Differential Revision: https://reviews.facebook.net/D15045
Summary:
Code was always compressing L0 files written by a memtable flush
when compression was enabled. Now this is done when
min_level_to_compress=0 for leveled compaction and when
universal_compaction_size_percent=-1 for universal compaction.
Task ID: #3416472
Blame Rev:
Test Plan:
ran db_bench with compression options
Revert Plan:
Database Impact:
Memcache Impact:
Other Notes:
EImportant:
- begin *PUBLIC* platform impact section -
Bugzilla: #
- end platform impact -
Reviewers: dhruba, igor, sdong
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D14757
Summary:
In addition to implementing OpenWithColumnFamilies, this diff also includes some minor changes:
* Changed all column family names from Slice() to std::string. The performance of column family name handling is not critical, and it's more convenient and cleaner to have names as std::strings
* Implemented ColumnFamilyOptions(const Options&) and DBOptions(const Options&)
* Added ColumnFamilyOptions to VersionSet::ColumnFamilyData. ColumnFamilyOptions are specified on OpenWithColumnFamilies() and CreateColumnFamily()
I will keep the diff in the Phabricator for a day or two and will push to the branch then. Feel free to comment even after the diff has been pushed.
Test Plan: Added a simple unit test
Reviewers: dhruba, haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D15033
Summary:
The way DBImpl::TEST_CompactRange() throttles down the number of bg compactions
can cause it to deadlock when CompactRange() is called concurrently from
multiple threads. Imagine a following scenario with only two threads
(max_background_compactions is 10 and bg_compaction_scheduled_ is initially 0):
1. Thread #1 increments bg_compaction_scheduled_ (to LargeNumber), sets
bg_compaction_scheduled_ to 9 (newvalue), schedules the compaction
(bg_compaction_scheduled_ is now 10) and waits for it to complete.
2. Thread #2 calls TEST_CompactRange(), increments bg_compaction_scheduled_
(now LargeNumber + 10) and waits on a cv for bg_compaction_scheduled_ to
drop to LargeNumber.
3. BG thread completes the first manual compaction, decrements
bg_compaction_scheduled_ and wakes up all threads waiting on bg_cv_.
Thread #1 runs, increments bg_compaction_scheduled_ by LargeNumber again
(now 2*LargeNumber + 9). Since that's more than LargeNumber + newvalue,
thread #2 also goes to sleep (waiting on bg_cv_), without resetting
bg_compaction_scheduled_.
This diff attempts to address the problem by introducing a new counter
bg_manual_only_ (when positive, MaybeScheduleFlushOrCompaction() will only
schedule manual compactions).
Test Plan:
I could pretty much consistently reproduce the deadlock with a program that
calls CompactRange(nullptr, nullptr) immediately after Write() from multiple
threads. This no longer happens with this patch.
Tests (make check) pass.
Reviewers: dhruba, igor, sdong, haobo
Reviewed By: igor
CC: leveldb
Differential Revision: https://reviews.facebook.net/D14799
Summary:
This diff provides basic implementations of CreateColumnFamily(), DropColumnFamily() and ListColumnFamilies(). It builds on top of https://reviews.facebook.net/D14733
It also includes a bug fix for DBImplReadOnly, where Get implementation would be redirected to DBImpl instead of DBImplReadOnly.
Test Plan: Added unit test
Reviewers: dhruba, haobo, kailiu
CC: leveldb
Differential Revision: https://reviews.facebook.net/D15021
Summary: this diff only replace the cases when we need to frequently create vector with small amount of entries. This diff doesn't aim to improve performance of a specific area, but more like a small scale test for the autovector and see how it works in real life.
Test Plan:
make check
I also ran the performance tests, however there is no performance gain/loss. All performance numbers are pretty much the same before/after the change.
Reviewers: dhruba, haobo, sdong, igor
CC: leveldb
Differential Revision: https://reviews.facebook.net/D14985
Summary: I made some cleanup while reading the source code in `db`. Most changes are about style, naming or C++ 11 new features.
Test Plan: ran `make check`
Reviewers: haobo, dhruba, sdong
CC: leveldb
Differential Revision: https://reviews.facebook.net/D15009
Summary:
We don't want two threads to clash if they concurrently call DisableFileDeletions() and EnableFileDeletions(). I'm adding a counter that will enable file deletions only after all DisableFileDeletions() calls have been negated with EnableFileDeletions().
However, we also don't want to break the old behavior, so I added a parameter force to EnableFileDeletions(). If force is true, we will still enable file deletions after every call to EnableFileDeletions(), which is what is happening now.
Test Plan: make check
Reviewers: dhruba, haobo, sanketh
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D14781
Summary:
Instead of locking and saving a DB state, we can cache a DB state and update it only when it changes. This change reduces lock contention and speeds up read operations on the DB.
Performance improvements are substantial, although there is some cost in no-read workloads. I ran the regression tests on my devserver and here are the numbers:
overwrite 56345 -> 63001
fillseq 193730 -> 185296
readrandom 771301 -> 1219803 (58% improvement!)
readrandom_smallblockcache 677609 -> 862850
readrandom_memtable_sst 710440 -> 1109223
readrandom_fillunique_random 221589 -> 247869
memtablefillrandom 105286 -> 92643
memtablereadrandom 763033 -> 1288862
Test Plan:
make asan_check
I am also running db_stress
Reviewers: dhruba, haobo, sdong, kailiu
Reviewed By: haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D14679
Summary:
<This diff is for Column Family branch>
Sharing some of the work I've done so far. This diff compiles and passes the tests.
The biggest change is in options.h - I broke down Options into two parts - DBOptions and ColumnFamilyOptions. DBOptions is DB-specific (env, create_if_missing, block_cache, etc.) and ColumnFamilyOptions is column family-specific (all compaction options, compresion options, etc.). Note that this does not break backwards compatibility at all.
Further, I created DBWithColumnFamily which inherits DB interface and adds new functions with column family support. Clients can transparently switch to DBWithColumnFamily and it will not break their backwards compatibility.
There are few methods worth checking out: ListColumnFamilies(), MultiNewIterator(), MultiGet() and GetSnapshot(). [GetSnapshot() returns the snapshot across all column families for now - I think that's what we agreed on]
Finally, I made small changes to WriteBatch so we are able to atomically insert data across column families.
Please provide feedback.
Test Plan: make check works, the code is backward compatible
Reviewers: dhruba, haobo, sdong, kailiu, emayanke
CC: leveldb
Differential Revision: https://reviews.facebook.net/D14445
Summary: It seems to be a decision tradeoff in current codes: we make a malloc for every Get() to reduce one malloc for a flush inside mutex. It takes about 5% of CPU time in readrandom tests. We might consider the tradeoff to be the other way around.
Test Plan: make all check
Reviewers: dhruba, haobo, igor
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D14697
Summary: To reduce mutex contention caused by DBImpl.NewInternalIterator(), in this function, move all the iteration creation works out of mutex, only leaving object ref and get.
Test Plan:
make all check
will run db_stress for a while too to make sure no problem.
Reviewers: haobo, dhruba, kailiu
Reviewed By: haobo
CC: igor, leveldb
Differential Revision: https://reviews.facebook.net/D14589
Summary: When deconstructing an iterator, no need to check obsolete file if it doesn't hold last reference of any version.
Test Plan: make all check
Reviewers: haobo, igor, dhruba, kailiu
Reviewed By: haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D14595
Summary: In get operations, merge_operands is only used in few cases. Lazily initialize it can reduce average latency in some cases
Test Plan: make all check
Reviewers: haobo, kailiu, dhruba
Reviewed By: haobo
CC: igor, nkg-, leveldb
Differential Revision: https://reviews.facebook.net/D14415
Conflicts:
db/db_impl.cc
db/memtable.cc
Summary:
creating new iterators of mem tables can be expensive. Move them out of mutex.
DBImpl::WriteLevel0Table()'s mems seems to be a local vector and is only used by flushing. memtables to flush are also immutable, so it should be safe to do so.
Test Plan: make all check
Reviewers: haobo, dhruba, kailiu
Reviewed By: dhruba
CC: igor, leveldb
Differential Revision: https://reviews.facebook.net/D14577
Conflicts:
db/db_impl.cc
Summary:
creating new iterators of mem tables can be expensive. Move them out of mutex.
DBImpl::WriteLevel0Table()'s mems seems to be a local vector and is only used by flushing. memtables to flush are also immutable, so it should be safe to do so.
Test Plan: make all check
Reviewers: haobo, dhruba, kailiu
Reviewed By: dhruba
CC: igor, leveldb
Differential Revision: https://reviews.facebook.net/D14577
Summary: So fflush() takes a lock which is heavyweight. I added flush_pending_, but more importantly, I removed LogFlush() from foreground threads.
Test Plan: ./db_test
Reviewers: dhruba, haobo
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D14535
Summary:
In this diff I present you BackupableDB v1. You can easily use it to backup your DB and it will do incremental snapshots for you.
Let's first describe how you would use BackupableDB. It's inheriting StackableDB interface so you can easily construct it with your DB object -- it will add a method RollTheSnapshot() to the DB object. When you call RollTheSnapshot(), current snapshot of the DB will be stored in the backup dir. To restore, you can just call RestoreDBFromBackup() on a BackupableDB (which is a static method) and it will restore all files from the backup dir. In the next version, it will even support automatic backuping every X minutes.
There are multiple things you can configure:
1. backup_env and db_env can be different, which is awesome because then you can easily backup to HDFS or wherever you feel like.
2. sync - if true, it *guarantees* backup consistency on machine reboot
3. number of snapshots to keep - this will keep last N snapshots around if you want, for some reason, be able to restore from an earlier snapshot. All the backuping is done in incremental fashion - if we already have 00010.sst, we will not copy it again. *IMPORTANT* -- This is based on assumption that 00010.sst never changes - two files named 00010.sst from the same DB will always be exactly the same. Is this true? I always copy manifest, current and log files.
4. You can decide if you want to flush the memtables before you backup, or you're fine with backing up the log files -- either way, you get a complete and consistent view of the database at a time of backup.
5. More things you can find in BackupableDBOptions
Here is the directory structure I use:
backup_dir/CURRENT_SNAPSHOT - just 4 bytes holding the latest snapshot
0, 1, 2, ... - files containing serialized version of each snapshot - containing a list of files
files/*.sst - sst files shared between snapshots - if one snapshot references 00010.sst and another one needs to backup it from the DB, it will just reference the same file
files/ 0/, 1/, 2/, ... - snapshot directories containing private snapshot files - current, manifest and log files
All the files are ref counted and deleted immediatelly when they get out of scope.
Some other stuff in this diff:
1. Added GetEnv() method to the DB. Discussed with @haobo and we agreed that it seems right thing to do.
2. Fixed StackableDB interface. The way it was set up before, I was not able to implement BackupableDB.
Test Plan:
I have a unittest, but please don't look at this yet. I just hacked it up to help me with debugging. I will write a lot of good tests and update the diff.
Also, `make asan_check`
Reviewers: dhruba, haobo, emayanke
Reviewed By: dhruba
CC: leveldb, haobo
Differential Revision: https://reviews.facebook.net/D14295
Summary: In get operations, merge_operands is only used in few cases. Lazily initialize it can reduce average latency in some cases
Test Plan: make all check
Reviewers: haobo, kailiu, dhruba
Reviewed By: haobo
CC: igor, nkg-, leveldb
Differential Revision: https://reviews.facebook.net/D14415
Summary: This would enable rocksdb users to get the db identity without depending on implementation details(storing that in IDENTITY file)
Test Plan: db/db_test (has identity checks)
Reviewers: dhruba, haobo, igor, kailiu
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D14463
Summary:
Let's get rid of TransformRep and it's children. We have confirmed that HashSkipListRep works better with multifeed, so there is no benefit to keeping this around.
This diff is mostly just deleting references to obsoleted functions. I also have a diff for fbcode that we'll need to push when we switch to new release.
I had to expose HashSkipListRepFactory in the client header files because db_impl.cc needs access to GetTransform() function for SanitizeOptions.
Test Plan: make check
Reviewers: dhruba, haobo, kailiu, sdong
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D14397
Summary:
I went through all remaining shared_ptrs and removed the ones that I found not-necessary. Only GenerateCachePrefix() is called fairly often, so don't expect much perf wins.
The ones that are left are accessed infrequently and I think we're fine with keeping them.
Test Plan: make asan_check
Reviewers: dhruba, haobo
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D14427
Summary:
The commit at 27bbef1180 had a memory leak
that was detected by valgrind. The memtable that has a refcount decrement
in MemTableList::InstallMemtableFlushResults was not freed.
Test Plan: valgrind ./db_test --leak-check=full
Reviewers: igor
Reviewed By: igor
CC: leveldb
Differential Revision: https://reviews.facebook.net/D14391
Summary:
Large memory allocations and frees are costly and best done outside the
db-mutex. The memtables are already allocated outside the db-mutex but
they were being freed while holding the db-mutex.
This patch frees obsolete memtables outside the db-mutex.
Test Plan:
make check
db_stress
Unit tests pass, I am in the process of running stress tests.
Reviewers: haobo, igor, emayanke
Reviewed By: haobo
CC: reconnect.grayhat, leveldb
Differential Revision: https://reviews.facebook.net/D14319
Summary: We need access to options for BackupableDB
Test Plan: make check
Reviewers: dhruba
Reviewed By: dhruba
CC: leveldb, reconnect.grayhat
Differential Revision: https://reviews.facebook.net/D14331
Summary: This is part of https://reviews.facebook.net/D14295 -- smaller diff that is easier to review
Test Plan: make asan_check
Reviewers: dhruba, haobo, emayanke
Reviewed By: emayanke
CC: leveldb, kailiu, reconnect.grayhat
Differential Revision: https://reviews.facebook.net/D14301
Summary:
All filesystem Io should be done outside the dbmutex. There was one place
when we have to roll the transaction log that we were creating the new log file
while holding the dbmutex.
I rearranged this code so that the act of creating the new transaction log
file is done without holding the dbmutex. I also allocate the new memtable
outside the dbmutex, this is important because creating the memtable
could be heavyweight.
Test Plan: make check and dbstress
Reviewers: haobo, igor
Reviewed By: haobo
CC: leveldb, reconnect.grayhat
Differential Revision: https://reviews.facebook.net/D14283
Summary: liveness of the statistics object is already ensured by the shared pointer in DB options. There's no reason to pass again shared pointer among internal functions. Raw pointer is sufficient and efficient.
Test Plan: make check
Reviewers: dhruba, MarkCallaghan, igor
Reviewed By: dhruba
CC: leveldb, reconnect.grayhat
Differential Revision: https://reviews.facebook.net/D14289
Summary:
Provide a framework to profile a query in detail to figure out latency bottleneck. Currently, in Get(), Put() and iterators, 2-3 simple timing is used. We can easily add more profile counters to the framework later.
Test Plan: Enable this profiling in seveal existing tests.
Reviewers: haobo, dhruba, kailiu, emayanke, vamsi, igor
CC: leveldb
Differential Revision: https://reviews.facebook.net/D14001
Conflicts:
table/merger.cc
Summary:
Provide a framework to profile a query in detail to figure out latency bottleneck. Currently, in Get(), Put() and iterators, 2-3 simple timing is used. We can easily add more profile counters to the framework later.
Test Plan: Enable this profiling in seveal existing tests.
Reviewers: haobo, dhruba, kailiu, emayanke, vamsi, igor
CC: leveldb
Differential Revision: https://reviews.facebook.net/D14001
Summary:
Previously we introduce a `flush_block_policy_factory` in Options, however, that options is strongly releated to Table based tables.
It will make more sense to move it to block based table's own factory class.
Test Plan: make check to pass existing tests
Reviewers: dhruba, haobo
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D14211
Summary:
The primary motivation of the changes is to make it easier to figure out the inside of the tables.
* rename "table stats" to "table properties" since now we have more than "integers" to store in the property block.
* Add filter block size to the basic table properties.
* Whenever a table is built, we'll log the table properties (the sample output is in Test Plan).
* Make an api to expose deleted keys.
Test Plan:
Passed all existing test. and the sample output of table stats:
==================================================================
Basic Properties
------------------------------------------------------------------
# data blocks: 1
# entries: 1
raw key size: 9
raw average key size: 9
raw value size: 9
raw average value size: 0
data block size: 25
index block size: 27
filter block size: 18
(estimated) table size: 70
filter policy: rocksdb.BuiltinBloomFilter
==================================================================
User collected properties: InternalKeyPropertiesCollector
------------------------------------------------------------------
kDeletedKeys: 1
==================================================================
Reviewers: dhruba, haobo
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D14187
Summary:
Previously in KeyMayExist(), if DB::Get() returns non-Status::OK(), we assumes key may not exist.
However, as if index block is not in block cache, Status::Incomplete() will return. Worse still, if
options::filter_delete is enabled, we may falsely ignore the "delete" operation:
https://github.com/facebook/rocksdb/blob/master/db/write_batch.cc#L217-L220
This diff fixes this bug and will let crash-test pass.
Test Plan:
Ran:
./db_stress --test_batches_snapshots=1 --ops_per_thread=1000000 --threads=32 --write_buffer_size=4194304 --destroy_db_initially=1 --reopen=0 --readpercent=5 --prefixpercent=45 --writepercent=35 --delpercent=5 --iterpercent=10 --db=/home/kailiu/local/newer --max_key=100000000 --disable_seek_compaction=0 --mmap_read=0 --block_size=16384 --cache_size=1048576 --open_files=500000 --verify_checksum=1 --sync=0 --disable_wal=0 --disable_data_sync=0 --target_file_size_base=2097152
--target_file_size_multiplier=2 --max_write_buffer_number=3 --max_background_compactions=20 --max_bytes_for_level_base=10485760 --filter_deletes=1
Previously we'll see crash happens very soon.
Reviewers: igor, dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D14115
Summary:
Created a unittest that verifies that automatic deletion performed by PurgeObsoleteFiles() works correctly.
Also, few small fixes on the logic part -- call version_set_->GetObsoleteFiles() in FindObsoleteFiles() instead of on some arbitrary positions.
Test Plan: Created a unit test
Reviewers: dhruba, haobo, nkg-
Reviewed By: haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D14079
Summary:
mac and our dev server has totally differnt definition of uint64_t, therefore fixing the warning in mac has actually made code in linux uncompileable.
Test Plan:
make clean && make -j32
Summary: One more fix! In some cases, our filenames start with "/". Apparently, env_ can't handle filenames with double //
Test Plan:
deletefile_test does not include this line in the LOG anymore:
2013/11/12-18:11:43.150149 7fe4a6fff700 RenameFile logfile #3 FAILED -- IO error: /tmp/rocksdbtest-3574/deletefile_test//000003.log: No such file or directory
Reviewers: dhruba, haobo
Reviewed By: haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D14055
Summary: The work to make sure mac os compiles rocksdb is not completed yet. But at least we can start cleaning some warnings captured only by g++ from mac os..
Test Plan: ran make in mac os
Reviewers: dhruba
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D14049
Summary:
@haobo's suggestions from https://reviews.facebook.net/D13827
Renaming some variables, deprecating purge_log_after_flush, changing for loop into auto for loop.
I have not implemented deleting objects outside of mutex yet because it would require a big code change - we would delete object in db_impl, which currently does not know anything about object because it's defined in version_edit.h (FileMetaData). We should do it at some point, though.
Test Plan: Ran deletefile_test
Reviewers: haobo
Reviewed By: haobo
CC: leveldb, haobo
Differential Revision: https://reviews.facebook.net/D14025
Summary: FindObsoleteFiles() has to be called before PurgeObsoleteFiles() because FindObsoleteFiles() sets manifest_file_number, log_number and prev_log_number to valid values.
Test Plan: deletefile_test now works
Reviewers: dhruba, emayanke, kailiu
Reviewed By: kailiu
CC: leveldb
Differential Revision: https://reviews.facebook.net/D13995
Summary:
Here's one solution we discussed on speeding up FindObsoleteFiles. Keep a set of all files in DBImpl and update the set every time we create a file. I probably missed few other spots where we create a file.
It might speed things up a bit, but makes code uglier. I don't really like it.
Much better approach would be to abstract all file handling to a separate class. Think of it as layer between DBImpl and Env. Having a separate class deal with file namings and deletion would benefit both code cleanliness (especially with huge DBImpl) and speed things up. It will take a huge effort to do this, though.
Let's discuss offline today.
Test Plan: Ran ./db_stress, verified that files are getting deleted
Reviewers: dhruba, haobo, kailiu, emayanke
Reviewed By: dhruba
Differential Revision: https://reviews.facebook.net/D13827
Summary: Allow block based table to configure the way flushing the blocks. This feature will allow us to add support for prefix-aligned block.
Test Plan: make check
Reviewers: dhruba, haobo, sdong, igor
Reviewed By: sdong
CC: leveldb
Differential Revision: https://reviews.facebook.net/D13875
Summary:
Added a new call LogFlush() that flushes the log contents to the OS buffers. We never call it with lock held.
We call it once for every Read/Write and often in compaction/flush process so the frequency should not be a problem.
Test Plan: db_test
Reviewers: dhruba, haobo, kailiu, emayanke
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D13935
Summary: Added a prefix_seek flag in ReadOptions to indicate that Seek is prefix aware(might not return data with different prefix), and also not bound to a specific prefix. Multiple Seeks and range scans can be invoked on the same iterator. If a specific prefix is specified, this flag will be ignored. Just a quick prototype that works for PrefixHashRep, the new lockless memtable could be easily extended with this support too.
Test Plan: test it on Leaf
Reviewers: dhruba, kailiu, sdong, igor
Reviewed By: igor
CC: leveldb
Differential Revision: https://reviews.facebook.net/D13929
Summary:
Archive cleaning will still happen every WAL_ttl seconds
but archived logs will be deleted only if archive size
is greater then a WAL_size_limit value.
Empty archived logs will be deleted evety WAL_ttl.
Test Plan:
1. Unit tests pass.
2. Benchmark.
Reviewers: emayanke, dhruba, haobo, sdong, kailiu, igor
Reviewed By: emayanke
CC: leveldb
Differential Revision: https://reviews.facebook.net/D13869
Summary:
strict essentially means that we MUST find the startsequence. Thus we should return if starteSequence is not found in the first file in case strict is set. This will take care of ending the iterator in case of permanent gaps due to corruptions in the log files
Also created NextImpl function that will have internal variable to distinguish whether Next is being called from StartSequence or by application.
Set NotFoudn::gaps status to give an indication of gaps happeneing.
Polished the inline documentation at various places
Test Plan:
* db_repl_stress test
* db_test relating to transaction log iterator
* fbcode/wormhole/rocksdb/rocks_log_iterator
* sigma production machine sigmafio032.prn1
Reviewers: dhruba
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D13689
Summary:
Rocksdb can now support a uncompressed block cache, or a compressed
block cache or both. Lookups first look for a block in the
uncompressed cache, if it is not found only then it is looked up
in the compressed cache. If it is found in the compressed cache,
then it is uncompressed and inserted into the uncompressed cache.
It is possible that the same block resides in the compressed cache
as well as the uncompressed cache at the same time. Both caches
have their own individual LRU policy.
Test Plan: Unit test case attached.
Reviewers: kailiu, sdong, haobo, leveldb
Reviewed By: haobo
CC: xjin, haobo
Differential Revision: https://reviews.facebook.net/D12675
Summary: This is to give application compaction filter a chance to access context information of a specific compaction run. For example, depending on whether a compaction goes through all data files, the application could do things differently.
Test Plan: make check
Reviewers: dhruba, kailiu, sdong
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D13683
Summary:
This patch is to address @haobo's comments on D13521:
1. rename Table to be TableReader and make its factory function to be GetTableReader
2. move the compression type selection logic out of TableBuilder but to compaction logic
3. more accurate comments
4. Move stat name constants into BlockBasedTable implementation.
5. remove some uncleaned codes in simple_table_db_test
Test Plan: pass test suites.
Reviewers: haobo, dhruba, kailiu
Reviewed By: haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D13785
Summary: This patch makes Table and TableBuilder a abstract class and make all the implementation of the current table into BlockedBasedTable and BlockedBasedTable Builder.
Test Plan: Make db_test.cc to work with block based table. Add a new test simple_table_db_test.cc where a different simple table format is implemented.
Reviewers: dhruba, haobo, kailiu, emayanke, vamsi
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D13521
Summary:
1. Added a new option that support user-defined table stats collection.
2. Added a deleted key stats collector in `utilities`
Test Plan:
Added a unit test for newly added code.
Also ran make check to make sure other tests are not broken.
Reviewers: dhruba, haobo
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D13491
Summary:
When a Put fails, it can leave database in a messy state. We don't want to pretend that everything is OK when it may not be. We fail every write following the failed one.
I added checks for corruption to DBImpl::Write(). Is there anywhere else I need to add them?
Test Plan: Corruption unit test.
Reviewers: dhruba, haobo, kailiu
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D13671
Summary:
This is to simplify rocksdb public APIs and improve the code quality.
Created an additional parameter to ParseFileName for log sub type and improved the code for deleting a wal file.
Wrote exhaustive unit-tests in delete_file_test
Unification of other redundant APIs can be taken up in a separate diff
Test Plan: Expanded delete_file test
Reviewers: dhruba, haobo, kailiu, sdong
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D13647
Summary:
Crash may occur during the flushes of more than two mem tables.
As the info log suggested, even when both were successfully flushed,
the recovery process still pick up one of the memtable's log for recovery.
This diff fix the problem by setting the correct "log number" in MANIFEST.
Test Plan: make test; deployed to leaf4 and make sure it doesn't result in crashes of this type.
Reviewers: haobo, dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D13659
Summary: as title
Test Plan: make check; ./perf_context_test
Reviewers: dhruba
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D13629
Summary:
Create a new type of file on startup if it doesn't already exist called DBID.
This will store a unique number generated from boost library's uuid header file.
The use-case is to identify the case of a db losing all its data and coming back up either empty or from an image(backup/live replica's recovery)
the key point to note is that DBID is not stored in a backup or db snapshot
It's preferable to use Boost for uuid because:
1) A non-standard way of generating uuid is not good
2) /proc/sys/kernel/random/uuid generates a uuid but only on linux environments and the solution would not be clean
3) c++ doesn't have any direct way to get a uuid
4) Boost is a very good library that was already having linkage in rocksdb from third-party
Note: I had to update the TOOLCHAIN_REV in build files to get latest verison of boost from third-party as the older version had a bug.
I had to put Wno-uninitialized in Makefile because boost-1.51 has an unitialized variable and rocksdb would not comiple otherwise. Latet open-source for boost is 1.54 but is not there in third-party. I have notified the concerned people in fbcode about it.
@kailiu : While releasing to third-party, an additional dependency will need to be created for boost in TARGETS file. I can help identify.
Test Plan:
Expand db_test to test 2 cases
1) Restarting db with Id file present - verify that no change to Id
2)Restarting db with Id file deleted - verify that a different Id is there after reopen
Also run make all check
Reviewers: dhruba, haobo, kailiu, sdong
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D13587
Summary:
This patch adds a option for universal compaction to allow us to only compress output files if the files compacted previously did not yet reach a specified ratio, to save CPU costs in some cases.
Compression is always skipped for flushing. This is because the size information is not easy to evaluate for flushing case. We can improve it later.
Test Plan:
add test
DBTest.UniversalCompactionCompressRatio1 and DBTest.UniversalCompactionCompressRatio12
Reviewers: dhruba, haobo
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D13467
Summary:
Enable background flush thread in this patch and fix unit tests with:
(1) After background flush, schedule a background compaction if condition satisfied;
(2) Fix a bug that if universal compaction is enabled and number of levels are set to be 0, compaction will not be automatically triggered
(3) Fix unit tests to wait for compaction to finish instead of flush, before checking the compaction results.
Test Plan: pass all unit tests
Reviewers: haobo, xjin, dhruba
Reviewed By: haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D13461
Summary:
* Logstore requests a valid change of reutrning an empty iterator and not an error in case of no log files.
* Changed the code to return the writebatch containing the sequence number requested from GetupdatesSince even if it lies in the middle. Earlier we used to return the next writebatch,. This also allows me oto guarantee that no files played upon by the iterator are redundant. I mean the starting log file has at least a sequence number >= the sequence number requested form GetupdatesSince.
* Cleaned up redundant logic in Iterator::Next and made a new function SeekToStartSequence for greater readability and maintainibilty.
* Modified a test in db_test accordingly
Please check the logic carefully and suggest improvements. I have a separate patch out for more improvements like restricting reader to read till written sequences.
Test Plan:
* transaction log iterator tests in db_test,
* db_repl_stress.
* rocks_log_iterator_test in fbcode/wormhole/rocksdb/test - 2 tests thriving on hacks till now can get simplified
* testing on the shadow setup for sigma with replication
Reviewers: dhruba, haobo, kailiu, sdong
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D13437
Summary:
So far we only have key/value pairs as well as bloom filter stored in the
sst file. It will be great if we are able to store more metadata about
this table itself, for example, the entry size, bloom filter name, etc.
This diff is the first step of this effort. It allows table to keep the
basic statistics mentioned in http://fburl.com/14995441, as well as
allowing writing user-collected stats to stats block.
After this diff, we will figure out the interface of how to allow user to collect their interested statistics.
Test Plan:
1. Added several unit tests.
2. Ran `make check` to ensure it doesn't break other tests.
Reviewers: dhruba, haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D13419
Summary: When I debug the unit test failures when enabling background flush thread, I feel the function names can be made clearer for people to understand. Also, if the names are fixed, in many places, some tests' bugs are obvious (and some of those tests are failing). This patch is to clean it up for future maintenance.
Test Plan: Run test suites.
Reviewers: haobo, dhruba, xjin
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D13431
Summary:
This careless error was causing ASSERT_OK(DestroyDB) to fail in db_test.
Basically .. was being returned as a child of db/archive and ParseFileName returned false on that,
but 'type' was set to LogFile from earlier and not reset. The return of ParseFileName was not being checked to delete the log file or not.
Test Plan: make all check
Reviewers: dhruba, haobo, xjin, kailiu, nkg-
Reviewed By: nkg-
CC: leveldb
Differential Revision: https://reviews.facebook.net/D13413
Summary: In some cases, you might not want to store the data log (write ahead log) files in the same dir as the sst files. An example use case is leaf, which stores sst files in tmpfs. And would like to save the log files in a separate dir (disk) to save memory.
Test Plan: make all. Ran db_test test. A few test failing. P2785018. If you guys don't see an obvious problem with the code, maybe somebody from the rocksdb team could help me debug the issue here. Running this on leaf worked well. I could see logs stored on disk, and deleted appropriately after compactions. Obviously this is only one set of options. The unit tests cover different options. Seems like I'm missing some edge cases.
Reviewers: dhruba, haobo, leveldb
CC: xinyaohu, sumeet
Differential Revision: https://reviews.facebook.net/D13239
Summary:
Change namespace from leveldb to rocksdb. This allows a single
application to link in open-source leveldb code as well as
rocksdb code into the same process.
Test Plan: compile rocksdb
Reviewers: emayanke
Reviewed By: emayanke
CC: leveldb
Differential Revision: https://reviews.facebook.net/D13287
Summary: This is useful to keep track of refreshes in transaction log iterator
Test Plan: make; db_stress --statistics=1 shows it
Reviewers: dhruba, haobo
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D13281
Summary: as title
Test Plan: make check
Reviewers: emayanke
Reviewed By: emayanke
CC: leveldb
Differential Revision: https://reviews.facebook.net/D13263
Summary: As title. This is just a quick hack and not ready for commit. fails a lot of unit test. I will test/debug it directly in ViewState shadow .
Test Plan: Try it in shadow test.
Reviewers: dhruba, xjin
CC: leveldb
Differential Revision: https://reviews.facebook.net/D12933
Summary:
We saw SIGSEGV when set options.num_levels=1 in universal compaction
style. Dug into this issue for a while, and finally found the root cause (thank Haobo for discussion).
Test Plan: Add new unit test. It throws SIGSEGV without this change. Also run "make all check".
Reviewers: haobo, dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D13251
Summary: A previous diff moved these outside of lock protected area. Moved back in now. Also moved tmp_batch_ update outside of lock protected area, as only the single write thread can access it.
Test Plan: make check
Reviewers: dhruba
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D13137
Summary: As title. The DB log file life cycle is tied up with the memtable it backs. Once the memtable is flushed to sst and committed, we should be able to delete the log file, without holding the mutex. This is part of the bigger change to avoid FindObsoleteFiles at runtime. It deals with log files. sst files will be dealt with later.
Test Plan: make check; db_bench
Reviewers: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D11709
Summary:
There is an config option called Options.min_write_buffer_number_to_merge
that specifies the minimum number of write buffers to merge in memory
before flushing to a file in L0. But in the the case when the db is
being closed, we should not be using this config, instead we should
flush whatever write buffers were available at that time.
Test Plan: Unit test attached.
Reviewers: haobo, emayanke
Reviewed By: haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D12717
Summary: So that replication can just download from wherever LogFile.Pathname is pointing them.
Test Plan: make all check;./db_repl_stress
Reviewers: dhruba, haobo
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D12609
Summary:
Add new command "change_compaction_style" to ldb tool. For
universal->level, it shows "nothing to do". For level->universal, it
compacts all files into a single one and moves the file to level 0.
Also add check for number of files at level 1+ when opening db with
universal compaction style.
Test Plan:
'make all check'. New unit test for internal convertion function. Also manully test various
cmd like:
./ldb change_compaction_style --old_compaction_style=0
--new_compaction_style=1 --db=/tmp/leveldbtest-3088/db_test
Reviewers: haobo, dhruba
Reviewed By: haobo
CC: vamsi, emayanke
Differential Revision: https://reviews.facebook.net/D12603
Summary:
The way counters/statistics are implemented in rocksdb demands that enum Tickers and TickerNameMap follow the same order, otherwise statistics exposed from fbcode/rocks get out-of-sync. 2 counters for prefix had violated this order and when I built counters for fbcode/mcrocksdb, statistics for sequence number were appearing out-of-sync.
The other change is to record sequence-number using setTickerCount only and not recordTick. This is because of difference in statistics as understood by rocks/utils which uses ServiceData::statistics function and rocksdb statistics. In rocksdb there is just 1 counter for a countername. But in ServiceData there are 4 independent buckets for every countername-Count, Sum, Average and Rate. SetTickerCount and RecordTick update the same variable in rocksdb but different buckets in ServiceData. Therefore, I had to choose one consistent function from RecordTick or SetTickerCount for sequence number in rocksdb. I chose SetTickerCount because the statistics object in options passed during rocksdb-open is user-dependent and SetTickerCount makes sense there.
There will be a corresponding diff to mcorcksdb in fbcode shortly.
Test Plan: make all check; check ticker value using fprintfs
Reviewers: dhruba
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D12669
Summary: db->DeleteFile calls ParseFileName to check name that was returned for sst file. Now, sst filename is returned using TableFileName which uses MakeFileName. This puts a / at the front of the name and ParseFileName doesn't like that. Changed ParseFileName to tolerate /s at the beginning. The test delet_file_test used to pass earlier because this behaviour of MakeFileName had been changed a while back to not return a / during which delete_file_test was checked in. But MakeFileName had to be reverted to add / at the front because GetLiveFiles used at many places outside rocksdb used the previous behaviour of MakeFileName.
Test Plan: make;./delete_filetest;make all check
Reviewers: dhruba, haobo, vamsi
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D12663
Summary:
The DeleteFile API was removing files inside the db-lock. This
is now changed to remove files outside the db-lock.
The GetLiveFilesMetadata() returns the smallest and largest
seqnuence number of each file as well.
Test Plan: deletefile_test
Reviewers: emayanke, haobo
Reviewed By: haobo
CC: leveldb
Maniphest Tasks: T63
Differential Revision: https://reviews.facebook.net/D12567
Summary: Let TransformRepFactory own the passed in transform. Also make it better encapsulated.
Test Plan: make valgrind_check;
Reviewers: dhruba, emayanke
Reviewed By: emayanke
CC: leveldb
Differential Revision: https://reviews.facebook.net/D12591
Summary:
If ReadOptions.non_blocking_io is set to true, then KeyMayExists
and Iterators will return data that is cached in RAM.
If the Iterator needs to do IO from storage to serve the data,
then the Iterator.status() will return Status::IsRetry().
Test Plan:
Enhanced unit test DBTest.KeyMayExist to detect if there were are IOs
issues from storage. Added DBTest.NonBlockingIteration to verify
nonblocking Iterations.
Reviewers: emayanke, haobo
Reviewed By: haobo
CC: leveldb
Maniphest Tasks: T63
Differential Revision: https://reviews.facebook.net/D12531
Summary:
As title. This is possible as tickers are atomic now.
db_bench on high qps in-memory muti-thread random get workload, showed ~5% throughput improvement.
Test Plan: make check; db_bench; db_stress
Reviewers: dhruba
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D12555
Summary: Replace include/leveldb with include/rocksdb.
Test Plan:
make clean; make check
make clean; make release
Differential Revision: https://reviews.facebook.net/D12489
Summary:
This patch adds three new MemTableRep's: UnsortedRep, PrefixHashRep, and VectorRep.
UnsortedRep stores keys in an std::unordered_map of std::sets. When an iterator is requested, it dumps the keys into an std::set and iterates over that.
VectorRep stores keys in an std::vector. When an iterator is requested, it creates a copy of the vector and sorts it using std::sort. The iterator accesses that new vector.
PrefixHashRep stores keys in an unordered_map mapping prefixes to ordered sets.
I also added one API change. I added a function MemTableRep::MarkImmutable. This function is called when the rep is added to the immutable list. It doesn't do anything yet, but it seems like that could be useful. In particular, for the vectorrep, it means we could elide the extra copy and just sort in place. The only reason I haven't done that yet is because the use of the ArenaAllocator complicates things (I can elaborate on this if needed).
Test Plan:
make -j32 check
./db_stress --memtablerep=vector
./db_stress --memtablerep=unsorted
./db_stress --memtablerep=prefixhash --prefix_size=10
Reviewers: dhruba, haobo, emayanke
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D12117
Summary: An api to query the level, key ranges, size etc for each SST file and an api to delete a specific file from the db and all associated state in the bookkeeping datastructures.
Notes: Editing the manifest version does not release the obsolete files right away. However deleting the file directly will mess up the iterator. We may need a more aggressive/timely file deletion api.
I have used std::unique_ptr - will switch to boost:: since this is external. thoughts?
Unit test is fragile right now as it expects the compaction at certain levels.
Test Plan: unittest
Reviewers: dhruba, vamsi, emayanke
CC: zshao, leveldb, haobo
Task ID: #
Blame Rev:
Test Plan:
- make all check;
- make release;
- make stringappend_test; ./stringappend_test
Reviewers: haobo, emayanke
Reviewed By: haobo
CC: leveldb, kailiu
Differential Revision: https://reviews.facebook.net/D12381
Summary: Also expanded class LogFile to have startSequene and FileSize and exposed it publicly
Test Plan: make all check
Reviewers: dhruba, haobo
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D12087
Summary: statistic for sequence number is needed by wormhole. setTickerCount is demanded for this statistic. I can't simply recordTick(max_sequence) when db recovers because the statistic iobject is owned by client and may/may not be reset during reopen. Eg. statistic is reset in mcrocksdb whereas it is not in db_stress. Therefore it is best to go with setTickerCount
Test Plan: ./db_stress ... --statistics=1 and observed expected sequence number
Reviewers: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D12327
Summary:
Minor fix to current codes, including: coding style, output format,
comments. No major logic change. There are only 2 real changes, please see my inline comments.
Test Plan: make all check
Reviewers: haobo, dhruba, emayanke
Differential Revision: https://reviews.facebook.net/D12297
Summary:
With Merge returning bool, it can keep failing silently(eg. While faling to fetch timestamp in TTL). We need to detect this through a rocksdb counter which can get bumped whenever Merge returns false. This will also be super-useful for the mcrocksdb-counter service where Merge may fail.
Added a counter NUMBER_MERGE_FAILURES and appropriately updated db/merge_helper.cc
I felt that it would be better to directly add counter-bumping in Merge as a default function of MergeOperator class but user should not be aware of this, so this approach seems better to me.
Test Plan: make all check
Reviewers: dnicholas, haobo, dhruba, vamsi
CC: leveldb
Differential Revision: https://reviews.facebook.net/D12129
Summary: Similar to v2 (db and table code understands prefixes), but use ReadOptions as in v3. Also, make the CreateFilter code faster and cleaner.
Test Plan: make db_test; export LEVELDB_TESTS=PrefixScan; ./db_test
Reviewers: dhruba
Reviewed By: dhruba
CC: haobo, emayanke
Differential Revision: https://reviews.facebook.net/D12027
Summary:
If we have same compaction filter for each compaction,
application cannot know about the different compaction processes.
Later on, we can put in more details in compaction filter for the
application to consume and use it according to its needs. For e.g. In
the universal compaction, we have a compaction process involving all the
files while others don't involve all the files. Applications may want to
collect some stats only when during full compaction.
Test Plan: run existing unit tests
Reviewers: haobo, dhruba
Reviewed By: dhruba
CC: xinyaohu, leveldb
Differential Revision: https://reviews.facebook.net/D12057
Summary:
The pre-existing code was purging a DeleteMarker if thay key did not
exist in deeper levels. But in the Universal Compaction Style, all
files are in Level0. For compaction runs that did not include the
earliest file, we were erroneously purging the DeleteMarkers.
The fix is to purge DeleteMarkers only if the compaction includes
the earlist file.
Test Plan: DBTest.Randomized triggers this code path.
Differential Revision: https://reviews.facebook.net/D12081
Summary:
Continue fixing existing unit tests for universal compaction. I have
tried to apply universal compaction to all unit tests those haven't
called ChangeOptions(). I left a few which are either apparently not
applicable to universal compaction (because they check files/keys/values
at level 1 or above levels), or apparently not related to compaction
(e.g., open a file, open a db).
I also add a new unit test for universal compaction.
Good news is I didn't see any bugs during this round.
Test Plan: Ran "make all check" yesterday. Has rebased and is rerunning
Reviewers: haobo, dhruba
Differential Revision: https://reviews.facebook.net/D12135
Summary:
This is the first step to fix unit tests and bugs for universal
compactiion. I added universal compaction option to ChangeOptions(), and
fixed all unit tests calling ChangeOptions(). Some of these tests
obviously assume more than 1 level and check file number/values in level
1 or above levels. I set kSkipUniversalCompaction for these tests.
The major bug I found is manual compaction with universal compaction never stops. I have put a fix for
it.
I have also set universal compaction as the default compaction and found
at least 20+ unit tests failing. I haven't looked into the details. The
next step is to check all unit tests without calling ChangeOptions().
Test Plan: make all check
Reviewers: dhruba, haobo
Differential Revision: https://reviews.facebook.net/D12051
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
Summary: This diff adds histogram stats for soft_rate_limit stalls. It also renames the old rate_limit stats to hard_rate_limit.
Test Plan: make -j32 check
Reviewers: dhruba, haobo, MarkCallaghan
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D12021
Summary:
This diff adds support for both soft and hard rate limiting. The following changes are included:
1) Options.rate_limit is renamed to Options.hard_rate_limit.
2) Options.rate_limit_delay_milliseconds is renamed to Options.rate_limit_delay_max_milliseconds.
3) Options.soft_rate_limit is added.
4) If the maximum compaction score is > hard_rate_limit and rate_limit_delay_max_milliseconds == 0, then writes are delayed by 1 ms at a time until the max compaction score falls below hard_rate_limit.
5) If the max compaction score is > soft_rate_limit but <= hard_rate_limit, then writes are delayed by 0-1 ms depending on how close we are to hard_rate_limit.
6) Users can disable 4 by setting hard_rate_limit = 0. They can add a limit to the maximum amount of time waited by setting rate_limit_delay_max_milliseconds > 0. Thus, the old behavior can be preserved by setting soft_rate_limit = 0, which is the default.
Test Plan:
make -j32 check
./db_stress
Reviewers: dhruba, haobo, MarkCallaghan
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D12003
Summary: Removed KeyMayExistImpl because KeyMayExist demanded Get like semantics now. Removed no_io from memtable and imm because we need the proper value now and shouldn't just stop when we see Merge in memtable. Added checks to block_cache. Updated documentation and unit-test
Test Plan: make all check;db_stress for 1 hour
Reviewers: dhruba, haobo
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D11853
Summary:
Currently, when a certain number of level0 files (level0_slowdown_writes_trigger) are present, RocksDB will slow down each write by 1ms. There is a second limit of level0 files at which RocksDB will stop writes altogether (level0_stop_writes_trigger).
This patch enables the user to supply a third parameter specifying the number of files at which Rocks will start slowing down writes (level0_start_slowdown_writes). When this number is reached, Rocks will slow down writes as a quadratic function of level0_slowdown_writes_trigger - num_level0_files.
For some workloads, this improves latency and throughput. I will post some stats momentarily in https://our.intern.facebook.com/intern/tasks/?t=2613384.
Test Plan:
make -j32 check
./db_stress
./db_bench
Reviewers: dhruba, haobo, MarkCallaghan, xjin
Reviewed By: xjin
CC: leveldb, xjin, zshao
Differential Revision: https://reviews.facebook.net/D11859
Summary:
Add an option for arena block size, default value 4096 bytes. Arena will allocate blocks with such size.
I am not sure about passing parameter to skiplist in the new virtualized framework, though I talked to Jim a bit. So add Jim as reviewer.
Test Plan:
new unit test, I am running db_test.
For passing paramter from configured option to Arena, I tried tests like:
TEST(DBTest, Arena_Option) {
std::string dbname = test::TmpDir() + "/db_arena_option_test";
DestroyDB(dbname, Options());
DB* db = nullptr;
Options opts;
opts.create_if_missing = true;
opts.arena_block_size = 1000000; // tested 99, 999999
Status s = DB::Open(opts, dbname, &db);
db->Put(WriteOptions(), "a", "123");
}
and printed some debug info. The results look good. Any suggestion for such a unit-test?
Reviewers: haobo, dhruba, emayanke, jpaton
Reviewed By: dhruba
CC: leveldb, zshao
Differential Revision: https://reviews.facebook.net/D11799
Summary: After my patch for stall histograms, there are redundant calls to NowMicros() by both the stop watches and DBImpl::MakeRoomForWrites. So I removed the redundant calls such that the information is gotten from the stopwatch.
Test Plan:
make clean
make -j32 check
Reviewers: dhruba, haobo, MarkCallaghan
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D11883
Summary: Previously, statistics are kept on how much time is spent on stalls of different types. This patch adds support for keeping number of stalls of each type. For example, instead of just reporting how many microseconds are spent waiting for memtables to be compacted, it will also report how many times a write stalled for that to occur.
Test Plan:
make -j32 check
./db_stress
# Not really sure what else should be done...
Reviewers: dhruba, MarkCallaghan, haobo
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D11841
Summary:
Revert "If disable wal is set, then batch commits are avoided" because
keeping the mutex while inserting into the skiplist means that readers
and writes are all serialized on the mutex.
Test Plan:
Reviewers:
CC:
Task ID: #
Blame Rev:
Summary: This diff virtualizes the skiplist interface so that users can provide their own implementation of a backing store for MemTables. Eventually, the backing store will be responsible for its own synchronization, allowing users (and us) to experiment with different lockless implementations.
Test Plan:
make clean
make -j32 check
./db_stress
Reviewers: dhruba, emayanke, haobo
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D11739
Summary:
rocksdb uses batch commit to write to transaction log. But if
disable wal is set, then writes to transaction log are anyways
avoided. In this case, there is not much value-add to batch things,
batching can cause unnecessary delays to Puts().
This patch avoids batching when disableWal is set.
Test Plan:
make check.
I am running db_stress now.
Reviewers: haobo
Reviewed By: haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D11763
Summary:
Introduced KeyMayExist checking during writebatch-delete and removed from Outer Delete API because it uses writebatch-delete.
Added code to skip getting Table from disk if not already present in table_cache.
Some renaming of variables.
Introduced KeyMayExistImpl which allows checking since specified sequence number in GetImpl useful to check partially written writebatch.
Changed KeyMayExist to not be pure virtual and provided a default implementation.
Expanded unit-tests in db_test to check appropriately.
Ran db_stress for 1 hour with ./db_stress --max_key=100000 --ops_per_thread=10000000 --delpercent=50 --filter_deletes=1 --statistics=1.
Test Plan: db_stress;make check
Reviewers: dhruba, haobo
Reviewed By: dhruba
CC: leveldb, xjin
Differential Revision: https://reviews.facebook.net/D11745
Summary: As title. This diff added an option reduce_level to CompactRange. When set to true, it will try to move the files back to the minimum level sufficient to hold the data set. Note that the default is set to true now, just to excerise it in all existing tests. Will set the default to false before check-in, for backward compatibility.
Test Plan: make check;
Reviewers: dhruba, emayanke
CC: leveldb
Differential Revision: https://reviews.facebook.net/D11553
Summary:
Wrote a new function in db_impl.c-CheckKeyMayExist that calls Get but with a new parameter turned on which makes Get return false only if bloom filters can guarantee that key is not in database. Delete calls this function and if the option- deletes_use_filter is turned on and CheckKeyMayExist returns false, the delete will be dropped saving:
1. Put of delete type
2. Space in the db,and
3. Compaction time
Test Plan:
make all check;
will run db_stress and db_bench and enhance unit-test once the basic design gets approved
Reviewers: dhruba, haobo, vamsi
Reviewed By: haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D11607
Summary: Replication logic would be simplifeid if we can guarantee that write sequence number is always contiguous, even if write failure occurs. Dhruba and I looked at the sequence number generation part of the code. It seems fixable. Note that if WAL was successful and insert into memtable was not, we would be in an unfortunate state. The approach in this diff is : IO error is expected and error status will be returned to client, sequence number will not be advanced; In-mem error is not expected and we panic.
Test Plan: make check; db_stress
Reviewers: dhruba, sheki
CC: leveldb
Differential Revision: https://reviews.facebook.net/D11439
Summary:
There is a new option called hybrid_mode which, when switched on,
causes HBase style compactions. Files from L0 are
compacted back into L0. This meat of this compaction algorithm
is in PickCompactionHybrid().
All files reside in L0. That means all files have overlapping
keys. Each file has a time-bound, i.e. each file contains a
range of keys that were inserted around the same time. The
start-seqno and the end-seqno refers to the timeframe when
these keys were inserted. Files that have contiguous seqno
are compacted together into a larger file. All files are
ordered from most recent to the oldest.
The current compaction algorithm starts to look for
candidate files starting from the most recent file. It continues to
add more files to the same compaction run as long as the
sum of the files chosen till now is smaller than the next
candidate file size. This logic needs to be debated
and validated.
The above logic should reduce write amplification to a
large extent... will publish numbers shortly.
Test Plan: dbstress runs for 6 hours with no data corruption (tested so far).
Differential Revision: https://reviews.facebook.net/D11289
Summary:
Merge multiple multiple memtables in memory before writing it
out to a file in L0.
There is a new config parameter min_write_buffer_number_to_merge
that specifies the number of write buffers that should be merged
together to a single file in storage. The system will not flush
wrte buffers to storage unless at least these many buffers have
accumulated in memory.
The default value of this new parameter is 1, which means that
a write buffer will be immediately flushed to disk as soon it is
ready.
Test Plan: make check
Differential Revision: https://reviews.facebook.net/D11241
Summary: This hopefully gives the right semantics to compaction filter. Will write a small wiki to explain the ideas.
Test Plan: make check; db_stress
Reviewers: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D11121
Summary:
Preliminary! Introduced the --use_multiget=1 and --keys_per_multiget=n
flags for db_bench. Also updated and tested the ReadRandom() method
to include an option to use multiget. By default,
keys_per_multiget=100.
Preliminary tests imply that multiget is at least 1.25x faster per
key than regular get.
Will continue adding Multiget for ReadMissing, ReadHot,
RandomWithVerify, ReadRandomWriteRandom; soon. Will also think
about ways to better verify benchmarks.
Test Plan:
1. make db_bench
2. ./db_bench --benchmarks=fillrandom
3. ./db_bench --benchmarks=readrandom --use_existing_db=1
--use_multiget=1 --threads=4 --keys_per_multiget=100
4. ./db_bench --benchmarks=readrandom --use_existing_db=1
--threads=4
5. Verify ops/sec (and 1000000 of 1000000 keys found)
Reviewers: haobo, MarkCallaghan, dhruba
Reviewed By: MarkCallaghan
CC: leveldb
Differential Revision: https://reviews.facebook.net/D11127
Summary:
This diff simplifies EnvOptions by treating it as POD, similar to Options.
- virtual functions are removed and member fields are accessed directly.
- StorageOptions is removed.
- Options.allow_readahead and Options.allow_readahead_compactions are deprecated.
- Unused global variables are removed: useOsBuffer, useFsReadAhead, useMmapRead, useMmapWrite
Test Plan: make check; db_stress
Reviewers: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D11175
Summary:
Without this files could be written out to a level greater than the maximum level possible and is the source of the segfaults that wormhole awas getting. The sequence of steps that was followed:
1. WriteLevel0Table was called when memtable was to be flushed for a file.
2. PickLevelForMemTableOutput was called to determine the level to which this file should be pushed.
3. PickLevelForMemTableOutput returned a wrong result because max_mem_compaction_level was equal to 2 even when num_levels was equal to 0.
The fix to re-initialize max_mem_compaction_level based on num_levels passed seems correct.
Test Plan: make all check; Also made a dummy file to mimic the wormhole-file behaviour which was causing the segfaults and found that the same segfault occurs without this change and not with this.
Reviewers: dhruba, haobo
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D11157
Summary:
Implemented the MultiGet operator which takes in a list of keys
and returns their associated values. Currently uses std::vector as its
container data structure. Otherwise, it works identically to "Get".
Test Plan:
1. make db_test ; compile it
2. ./db_test ; test it
3. make all check ; regress / run all tests
4. make release ; (optional) compile with release settings
Reviewers: haobo, MarkCallaghan, dhruba
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D10875
Summary: Added the 'score' column to the compaction stats output, which shows the level total size devided by level target size. Could be useful when monitoring compaction decisions...
Test Plan: make check; db_bench
Reviewers: dhruba
CC: leveldb, MarkCallaghan
Differential Revision: https://reviews.facebook.net/D11025
Summary:
This diff adds an option to specify whether PTHREAD_MUTEX_ADAPTIVE_NP will be enabled for the rocksdb single big kernel lock. db_bench also have this option now.
Quickly tested 8 thread cpu bound 100 byte random read.
No fast mutex: ~750k/s ops
With fast mutex: ~880k/s ops
Test Plan: make check; db_bench; db_stress
Reviewers: dhruba
CC: MarkCallaghan, leveldb
Differential Revision: https://reviews.facebook.net/D11031
Summary: a new option block_size_deviation is added.
Test Plan: run db_test and db_bench
Reviewers: dhruba, haobo
Reviewed By: haobo
Differential Revision: https://reviews.facebook.net/D10821
Summary: MaybeDumpStats was causing lock problem
Test Plan: make check; db_stress
Reviewers: dhruba
Reviewed By: dhruba
Differential Revision: https://reviews.facebook.net/D10935
Summary:
Added an option stats_dump_period_sec to dump leveldb.stats to LOG periodically for diagnosis.
By defauly, it's set to a very big number 3600 (1 hour).
Test Plan: make check;
Reviewers: dhruba
Reviewed By: dhruba
CC: leveldb, zshao
Differential Revision: https://reviews.facebook.net/D10761
Summary: There was an artifical limit on the size of the write buffer size.
Test Plan: make check
Reviewers: haobo
Reviewed By: haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D10911
Summary:
Currently, with paranoid_check on, DB::Open will fail on any log read error on recovery.
If client is ok with losing most recent updates, we could simply skip those errors.
However, it's important to introduce an additional flag, so that paranoid_check can
still guard against more serious problems.
Test Plan: make check; db_stress
Reviewers: dhruba, emayanke
Reviewed By: emayanke
CC: leveldb, emayanke
Differential Revision: https://reviews.facebook.net/D10869
Summary:
Make stop watch a simple implementation, instead of subclass of a virtual class
Allocate stop watches off the stack instead of heap.
Code is more terse now.
Test Plan: make all check, db_bench with --statistics=1
Reviewers: haobo, dhruba
Reviewed By: haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D10809
Summary:
This diff replaces compaction_filter_args and CompactionFilter with a single compaction_filter parameter. It gives CompactionFilter better encapsulation and a similar look to Comparator and MergeOpertor, which improves consistency of the overall interface.
The change is not backward compatible. Nevertheless, the two references in fbcode are not in production yet.
Test Plan: make check
Reviewers: dhruba
Reviewed By: dhruba
CC: leveldb, zshao
Differential Revision: https://reviews.facebook.net/D10773
Summary:
Currently, compaction filter is run on internal key older than the oldest snapshot, which is incorrect.
Compaction filter should really be run on the most recent internal key when there is no external snapshot.
Test Plan: make check; db_stress
Reviewers: dhruba
Reviewed By: dhruba
Differential Revision: https://reviews.facebook.net/D10641
Summary:
WAL files are moved to archive directory and clear only at DB::Open.
Can lead to a lot of space consumption in a Database. Added logic to periodically clear Archive Directory too.
Test Plan: make all check + add unit test
Reviewers: dhruba, heyongqiang
Reviewed By: heyongqiang
CC: leveldb
Differential Revision: https://reviews.facebook.net/D10617
Summary:
This diff introduces a new Merge operation into rocksdb.
The purpose of this review is mostly getting feedback from the team (everyone please) on the design.
Please focus on the four files under include/leveldb/, as they spell the client visible interface change.
include/leveldb/db.h
include/leveldb/merge_operator.h
include/leveldb/options.h
include/leveldb/write_batch.h
Please go over local/my_test.cc carefully, as it is a concerete use case.
Please also review the impelmentation files to see if the straw man implementation makes sense.
Note that, the diff does pass all make check and truly supports forward iterator over db and a version
of Get that's based on iterator.
Future work:
- Integration with compaction
- A raw Get implementation
I am working on a wiki that explains the design and implementation choices, but coding comes
just naturally and I think it might be a good idea to share the code earlier. The code is
heavily commented.
Test Plan: run all local tests
Reviewers: dhruba, heyongqiang
Reviewed By: dhruba
CC: leveldb, zshao, sheki, emayanke, MarkCallaghan
Differential Revision: https://reviews.facebook.net/D9651
Summary:
- don't see a point exposing table.h to the public.
- fixed make clean to remove also *.d files.
Test Plan: make check; db_stress
Reviewers: dhruba, heyongqiang
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D10479
Summary:
- removed the compaction_filter_value from the callback interface. Restrict compaction filter to purging values.
- modify some comments to reflect curent status.
Test Plan: make check
Reviewers: dhruba
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D10335
Summary: A better error message. A local change. Did not look at other places where this could be done.
Test Plan: compile
Reviewers: dhruba, MarkCallaghan
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D10251
Summary:
FindObsoleteFiles was slow, holding the single big lock, resulted in bad p99 behavior.
Didn't profile anything, but several things could be improved:
1. VersionSet::AddLiveFiles works with std::set, which is by itself slow (a tree).
You also don't know how many dynamic allocations occur just for building up this tree.
switched to std::vector, also added logic to pre-calculate total size and do just one allocation
2. Don't see why env_->GetChildren() needs to be mutex proteced, moved to PurgeObsoleteFiles where
mutex could be unlocked.
3. switched std::set to std:unordered_set, the conversion from vector is also inside PurgeObsoleteFiles
I have a feeling this should pretty much fix it.
Test Plan: make check; db_stress
Reviewers: dhruba, heyongqiang, MarkCallaghan
Reviewed By: dhruba
CC: leveldb, zshao
Differential Revision: https://reviews.facebook.net/D10197
Summary: using unique_ptr to have automatic delete for probableWALfiles in db_impl.cc
Test Plan: make
Reviewers: sheki, dhruba
Reviewed By: sheki
CC: leveldb
Differential Revision: https://reviews.facebook.net/D10083
Summary:
The segfault was happening because the program was unable to open a new
sst file (as part of the compaction) because the process ran out of
file descriptors.
The fix is to check the return status of the file creation before taking
any other action.
Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x7fabf03f9700 (LWP 29904)]
leveldb::DBImpl::OpenCompactionOutputFile (this=this@entry=0x7fabf9011400, compact=compact@entry=0x7fabf741a2b0) at db/db_impl.cc:1399
1399 db/db_impl.cc: No such file or directory.
(gdb) where
Test Plan: make check
Reviewers: MarkCallaghan, sheki
Reviewed By: MarkCallaghan
CC: leveldb
Differential Revision: https://reviews.facebook.net/D10101
Summary:
Transaction Log Iterator did not move to the next file in the series if there was a write batch at the end of the currentFile.
The solution is if the last seq no. of the current file is < RequestedSeqNo. Assume the first seqNo. of the next file has to satisfy the request.
Also major refactoring around the code. Moved opening the logreader to a seperate function, got rid of goto.
Test Plan: added a unit test for it.
Reviewers: dhruba, heyongqiang
Reviewed By: heyongqiang
CC: leveldb, emayanke
Differential Revision: https://reviews.facebook.net/D10029
Summary:
During recovery, last_updated_manifest number was not set if there were no records in the Write-ahead log.
Now check for the recovered manifest also and set last_updated_manifest file to the max value.
Test Plan: unit test
Reviewers: heyongqiang
Reviewed By: heyongqiang
CC: leveldb
Differential Revision: https://reviews.facebook.net/D9891
Summary:
If a class owns an object:
- If the object can be null => use a unique_ptr. no delete
- If the object can not be null => don't even need new, let alone delete
- for runtime sized array => use vector, no delete.
Test Plan: make check
Reviewers: dhruba, heyongqiang
Reviewed By: heyongqiang
CC: leveldb, zshao, sheki, emayanke, MarkCallaghan
Differential Revision: https://reviews.facebook.net/D9783
Summary:
RocksDB does a binary search to look at the files which might contain the requested sequence number at the call GetUpdatesSince.
There was a bug in the binary search => when the file pointed by the middle index of bsearch was empty/corrupt it needst to resize the vector and update indexes.
This now fixes that.
Test Plan: existing unit tests pass.
Reviewers: heyongqiang, dhruba
Reviewed By: heyongqiang
CC: leveldb
Differential Revision: https://reviews.facebook.net/D9777
Summary:
If the vector returned by GetUpdatesSince is empty, it is still returned to the
user. This causes it throw an std::range error.
The probable file list is checked and it returns an IOError status instead of OK now.
Test Plan: added a unit test.
Reviewers: dhruba, heyongqiang
Reviewed By: heyongqiang
CC: leveldb
Differential Revision: https://reviews.facebook.net/D9771
Summary:
Use non mmapd files for Write-Ahead log.
Earlier use of MMaped files. made the log iterator read ahead and miss records.
Now the reader and writer will point to the same physical location.
There is no perf regression :
./db_bench --benchmarks=fillseq --db=/dev/shm/mmap_test --num=$(million 20) --use_existing_db=0 --threads=2
with This diff :
fillseq : 10.756 micros/op 185281 ops/sec; 20.5 MB/s
without this dif :
fillseq : 11.085 micros/op 179676 ops/sec; 19.9 MB/s
Test Plan: unit test included
Reviewers: dhruba, heyongqiang
Reviewed By: heyongqiang
CC: leveldb
Differential Revision: https://reviews.facebook.net/D9741
Summary: rocksdb uses a single global lock to protect in memory metadata. We should minimize the mutex protected code section to increase the effective parallelism of the program. See https://our.intern.facebook.com/intern/tasks/?t=2218928
Test Plan:
make check
db_bench
Reviewers: dhruba, heyongqiang
CC: zshao, leveldb
Differential Revision: https://reviews.facebook.net/D9705
Summary:
The events that trigger compaction:
* opening the database
* Get -> only if seek compaction is not disabled and other checks are true
* MakeRoomForWrite -> when memtable is full
* BackgroundCall ->
If the background thread is about to do a compaction run, it schedules
a new background task to trigger a possible compaction. This will cause
additional background threads to find and process other compactions that
can run concurrently.
Test Plan: ran db_bench with overwrite and readonly alternatively.
Reviewers: sheki, MarkCallaghan
Reviewed By: sheki
CC: leveldb
Differential Revision: https://reviews.facebook.net/D9579
Summary:
This patch allows an application to specify whether to use bufferedio,
reads-via-mmaps and writes-via-mmaps per database. Earlier, there
was a global static variable that was used to configure this functionality.
The default setting remains the same (and is backward compatible):
1. use bufferedio
2. do not use mmaps for reads
3. use mmap for writes
4. use readaheads for reads needed for compaction
I also added a parameter to db_bench to be able to explicitly specify
whether to do readaheads for compactions or not.
Test Plan: make check
Reviewers: sheki, heyongqiang, MarkCallaghan
Reviewed By: sheki
CC: leveldb
Differential Revision: https://reviews.facebook.net/D9429
Summary: Makefile had options to ignore sign-comparisons and unused-parameters, which should be there. Also fixed the specific errors in the code-base
Test Plan: make
Reviewers: chip, dhruba
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D9531
Summary:
Add --benchmarks=levelstats option to report per-level stats (#files, #bytes)
Change readwhilewriting test to report response time for writes but exclude
them from the stats merged by all threads.
Prevent "NaN" in stats output by preventing division by 0.
Remove "o" file I committed by mistake.
Task ID: #
Blame Rev:
Test Plan:
make check
Revert Plan:
Database Impact:
Memcache Impact:
Other Notes:
EImportant:
- begin *PUBLIC* platform impact section -
Bugzilla: #
- end platform impact -
Reviewers: dhruba
Reviewed By: dhruba
Differential Revision: https://reviews.facebook.net/D9513
Summary:
Rocksdb can create 0 sized log files when it is opened and closed without any operations.
The GetUpdatesSince fails currently if there is a log file of size zero.
This diff fixes this. If there is a log file is 0, it is removed form the probable_file_list
Test Plan: unit test
Reviewers: dhruba, heyongqiang
Reviewed By: heyongqiang
CC: leveldb
Differential Revision: https://reviews.facebook.net/D9507
Summary:
If there is an error while writing an edit to the manifest file, the manifest
file is closed and reopened to check if the edit made it in. However, if the
re-opening of the manifest is unsuccessful and options.paranoid_checks is set
t true, then the db refuses to accept new puts, effectively putting the db
in readonly mode.
In a future diff, I would like to make the default value of paranoid_check
to true.
Test Plan: make check
Reviewers: sheki
Reviewed By: sheki
CC: leveldb
Differential Revision: https://reviews.facebook.net/D9201
Summary:
Store the last flushed, seq no. in db_impl. Check against it in
transaction Log iterator. Do not attempt to read ahead if we do not know
if the data is flushed completely.
Does not work if flush is disabled. Any ideas on fixing that?
* Minor change, iter->Next is called the first time automatically for
* the first time.
Test Plan:
existing test pass.
More ideas on testing this?
Planning to run some stress test.
Reviewers: dhruba, heyongqiang
CC: leveldb
Differential Revision: https://reviews.facebook.net/D9087
Summary:
The compaction process zeros out sequence numbers if the output is
part of the bottommost level.
The Slice is supposed to refer to an immutable data buffer. The
merger that implements the priority queue while reading kvs as
the input of a compaction run reies on this fact. The bug was that
were updating the sequence number of a record in-place and that was
causing suceeding invocations of the merger to return kvs in
arbitrary order of sequence numbers.
The fix is to copy the key to a local memory buffer before setting
its seqno to 0.
Test Plan:
Set Options.purge_redundant_kvs_while_flush = false and then run
db_stress --ops_per_thread=1000 --max_key=320
Reviewers: emayanke, sheki
Reviewed By: emayanke
CC: leveldb
Differential Revision: https://reviews.facebook.net/D9147
Summary:
This adds the rate_delay_limit_milliseconds option to make the delay
configurable in MakeRoomForWrite when the max compaction score is too high.
This delay is called the Ln slowdown. This change also counts the Ln slowdown
per level to make it possible to see where the stalls occur.
From IO-bound performance testing, the Level N stalls occur:
* with compression -> at the largest uncompressed level. This makes sense
because compaction for compressed levels is much
slower. When Lx is uncompressed and Lx+1 is compressed
then files pile up at Lx because the (Lx,Lx+1)->Lx+1
compaction process is the first to be slowed by
compression.
* without compression -> at level 1
Task ID: #1832108
Blame Rev:
Test Plan:
run with real data, added test
Revert Plan:
Database Impact:
Memcache Impact:
Other Notes:
EImportant:
- begin *PUBLIC* platform impact section -
Bugzilla: #
- end platform impact -
Reviewers: dhruba
Reviewed By: dhruba
Differential Revision: https://reviews.facebook.net/D9045
Summary:
Rocks accumulates recent writes and deletes in the in-memory memtable.
When the memtable is full, it writes the contents on the memtable to
a file in L0.
This patch removes redundant records at the time of the flush. If there
are multiple versions of the same key in the memtable, then only the
most recent one is dumped into the output file. The purging of
redundant records occur only if the most recent snapshot is earlier
than the earliest record in the memtable.
Should we switch on this feature by default or should we keep this feature
turned off in the default settings?
Test Plan: Added test case to db_test.cc
Reviewers: sheki, vamsi, emayanke, heyongqiang
Reviewed By: sheki
CC: leveldb
Differential Revision: https://reviews.facebook.net/D8991
Summary:
There was an artifical limit of 50K files per database. This is
insifficient if the database is 1 TB in size and each file is 2 MB.
Test Plan: make check
Reviewers: sheki, emayanke
Reviewed By: emayanke
CC: leveldb
Differential Revision: https://reviews.facebook.net/D8919
Summary: just record time consumed in compaction
Test Plan: compile
Reviewers: dhruba
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D8781
Summary:
* Counters for bytes read and write.
as a part of this diff, I want to=>
* Measure compaction times. @dhruba can you point which function, should
* I time to get Compaction-times. Was looking at CompactRange.
Test Plan: db_test
Reviewers: dhruba, emayanke
CC: leveldb
Differential Revision: https://reviews.facebook.net/D8763
Summary:
* Introduce is histogram in statistics.h
* stop watch to measure time.
* introduce two timers as a poc.
Replaced NULL with nullptr to fight some lint errors
Should be useful for google.
Test Plan:
ran db_bench and check stats.
make all check
Reviewers: dhruba, heyongqiang
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D8637
Summary:
The sequence numbers in each record eat up plenty of space on storage.
The optimization zeroes out sequence numbers on kvs in the Lmax
layer that are earlier than the earliest snapshot.
Test Plan: Unit test attached.
Differential Revision: https://reviews.facebook.net/D8619
Summary:
flush_on_destroy has a default value of false and the memtable is flushed
in the dbimpl-destructor only when that is set to true. Because we want the memtable to be flushed everytime that
the destructor is called(db is closed) and the cases where we work with the memtable only are very less
it is a good idea to give this a default value of true. Thus the put from ldb
wil have its data flushed to disk in the destructor and the next Get will be able to
read it when opened with OpenForReadOnly. The reason that ldb could read the latest value when
the db was opened in the normal Open mode is that the Get from normal Open first reads
the memtable and directly finds the latest value written there and the Get from OpenForReadOnly
doesn't have access to the memtable (which is correct because all its Put/Modify) are disabled
Test Plan: make all; ldb put and get and scans
Reviewers: dhruba, heyongqiang, sheki
Reviewed By: heyongqiang
CC: kosievdmerwe, zshao, dilipj, kailiu
Differential Revision: https://reviews.facebook.net/D8631
Summary:
* Add a SplitByTTLLogger to enable this feature. In this diff I implemented generalized AutoSplitLoggerBase class to simplify the
development of such classes.
* Refactor the existing AutoSplitLogger and fix several bugs.
Test Plan:
* Added a unit tests for different types of "auto splitable" loggers individually.
* Tested the composited logger which allows the log files to be splitted by both TTL and log size.
Reviewers: heyongqiang, dhruba
Reviewed By: heyongqiang
CC: zshao, leveldb
Differential Revision: https://reviews.facebook.net/D8037
Summary:
Previously, if you opened a db with num_levels set lower than
the database, you received the unhelpful message "Corruption:
VersionEdit: new-file entry." Now you get a more verbose message
describing the issue.
Also, fix handling of compression_levels (both the run-over-the-end
issue and the memory management of it).
Lastly, unique_ptr'ify a couple of minor calls.
Test Plan: make check
Reviewers: dhruba
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D8151
Summary:
We continually rebuilt build_version.c because we put the
current date into it, but that's what __DATE__ already is. This makes
builds faster.
This also fixes an issue with 'make clean FOO' not working properly.
Also tweak the build rules to be more consistent, always have warnings,
and add a 'make release' rule to handle flags for release builds.
Test Plan: make, make clean
Reviewers: dhruba
Reviewed By: dhruba
Differential Revision: https://reviews.facebook.net/D8139
Summary:
On some filesystems, pre-allocation can be a considerable
amount of space. xfs in our production environment pre-allocates by
1GB, for instance. By using fallocate to inform the kernel of our
expected file sizes, we eliminate this wasteage (that isn't recovered
until the file is closed which, in the case of LOG files, can be a
considerable amount of time).
Test Plan:
created an xfs loopback filesystem, mounted with
allocsize=4M, and ran db_stress. LOG file without this change was 4M,
and with it it was 128k then grew to normal size.
Reviewers: dhruba
Reviewed By: dhruba
CC: adsharma, leveldb
Differential Revision: https://reviews.facebook.net/D7953
Summary:
Replace manual memory management with std::unique_ptr in a
number of places; not exhaustive, but this fixes a few leaks with file
handles as well as clarifies semantics of the ownership of file handles
with log classes.
Test Plan: db_stress, make check
Reviewers: dhruba
Reviewed By: dhruba
CC: zshao, leveldb, heyongqiang
Differential Revision: https://reviews.facebook.net/D8043
Summary:
Found issues with `db_test` and `db_stress` when running valgrind.
`DBImpl` had an issue where if an compaction failed then it will use the uninitialised file size of an output file is used. This manifested as the final call to output to the log in `DoCompactionWork()` branching on uninitialized memory (all the way down in printf's innards).
Test Plan:
Ran `valgrind --track_origins=yes ./db_test` and `valgrind ./db_stress` to see if issues disappeared.
Ran `make check` to see if there were no regressions.
Reviewers: vamsi, dhruba
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D8001
Summary:
Check in LogAndApply if the file size is more than the limit set in
Options.
Things to consider : will this be expensive?
Test Plan: make all check. Inputs on a new unit test?
Reviewers: dhruba
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D7701
Summary:
Specific changes:
1) Turn on -Werror so all warnings are errors
2) Fix some warnings the above now complains about
3) Add proper dependency support so changing a .h file forces a .c file
to rebuild
4) Automatically use fbcode gcc on any internal machine rather than
whatever system compiler is laying around
5) Fix jemalloc to once again be used in the builds (seemed like it
wasn't being?)
6) Fix issue where 'git' would fail in build_detect_version because of
LD_LIBRARY_PATH being set in the third-party build system
Test Plan:
make, make check, make clean, touch a header file, make sure
rebuild is expected
Reviewers: dhruba
Reviewed By: dhruba
Differential Revision: https://reviews.facebook.net/D7887
Summary: Found some issues running Valgrind on `db_test` (there are still some outstanding ones) and fixed them.
Test Plan:
make check
ran `valgrind ./db_test` and saw that errors no longer occur
Reviewers: dhruba, vamsi, emayanke, sheki
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D7803
Summary:
Changed CreateDir() to CreateDirIfMissing() so a directory that already exists now causes and error.
Fixed CreateDirIfMissing() and added Env.DirExists()
Test Plan:
make check to test for regessions
Ran the following to test if the error message is not about lock files not existing
./db_bench --db=dir/testdb
After creating a file "testdb", ran the following to see if it failed with sane error message:
./db_bench --db=testdb
Reviewers: dhruba, emayanke, vamsi, sheki
Reviewed By: emayanke
CC: leveldb
Differential Revision: https://reviews.facebook.net/D7707
Summary:
Leveldb has an api OpenForReadOnly() that opens the database
in readonly mode. This call had an option to not process the
transaction log. This patch removes this option and always
processes all transactions that had been committed. It has
been done in such a way that it does not create/write to
any new files in the process. The invariant of "no-writes"
to the leveldb data directory is still true.
This enhancement allows multiple threads to open the same database
in readonly mode and access all trancations that were committed right
upto the OpenForReadOnly call.
I changed the public API to match the new semantics because
there are no users who are currently using this api.
Test Plan: make clean check
Reviewers: sheki
Reviewed By: sheki
CC: leveldb
Differential Revision: https://reviews.facebook.net/D7479
Summary:
1. The OpenForReadOnly() call should not lock the db. This is useful
so that multiple processes can open the same database concurrently
for reading.
2. GetUpdatesSince should not error out if the archive directory
does not exist.
3. A new constructor for WriteBatch that can takes a serialized
string as a parameter of the constructor.
Test Plan: make clean check
Reviewers: sheki
Reviewed By: sheki
CC: leveldb
Differential Revision: https://reviews.facebook.net/D7449
Summary:
Added kMetaDatabase for meta-databases in db/filename.h along with supporting
fuctions.
Fixed switch in DBImpl so that it also handles kMetaDatabase.
Fixed DestroyDB() that it can handle destroying meta-databases.
Test Plan: make check
Reviewers: sheki, emayanke, vamsi, dhruba
Reviewed By: dhruba
Differential Revision: https://reviews.facebook.net/D7245
Summary:
C tests would fail sometimes as DestroyDB would return a Failure Status
message when deleting an archival directory which was not created
(WAL_ttl_seconds = 0).
Fix: Ignore the Status returned on Deleting Archival Directory.
Test Plan: * make check
Reviewers: dhruba, emayanke
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D7395
Summary:
* Fixed implementation bug in Binary_Searvch introduced in https://reviews.facebook.net/D7119
* Binary search is also overflow safe.
* Delete archive log files and archive dir during DestroyDB
Test Plan: make check
Reviewers: dhruba
CC: kosievdmerwe, emayanke
Differential Revision: https://reviews.facebook.net/D7263
Summary:
Implement a interface to retrieve the most current transaction
id from the database.
Test Plan: Added unit test.
Reviewers: sheki
Reviewed By: sheki
CC: leveldb
Differential Revision: https://reviews.facebook.net/D7269
Summary:
filename.h has functions to do similar things.
Moving code away from db_impl.cc
Test Plan: make check
Reviewers: dhruba
Reviewed By: dhruba
Differential Revision: https://reviews.facebook.net/D7251
Summary:
How it works:
* GetUpdatesSince takes a SequenceNumber.
* A LogFile with the first SequenceNumber nearest and lesser than the requested Sequence Number is found.
* Seek in the logFile till the requested SeqNumber is found.
* Return an iterator which contains logic to return record's one by one.
Test Plan:
* Test case included to check the good code path.
* Will update with more test-cases.
* Feedback required on test-cases.
Reviewers: dhruba, emayanke
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D7119
Summary:
A compaction is picked based on its score. It is useful to
print the compaction score in the LOG because it aids in
debugging. If one looks at the logs, one can find out why
a compaction was preferred over another.
Test Plan: make clean check
Differential Revision: https://reviews.facebook.net/D7137
Summary:
Create a directory "archive" in the DB directory.
During DeleteObsolteFiles move the WAL files (*.log) to the Archive directory,
instead of deleting.
Test Plan: Created a DB using DB_Bench. Reopened it. Checked if files move.
Reviewers: dhruba
Reviewed By: dhruba
Differential Revision: https://reviews.facebook.net/D6975
Summary:
Scripted and removed all trailing spaces and converted all tabs to
spaces.
Also fixed other lint errors.
All lint errors from this point of time should be taken seriously.
Test Plan: make all check
Reviewers: dhruba
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D7059
Summary:
LevelDB should delete almost-new keys when a long-open snapshot exists.
The previous behavior is to keep all versions that were created after the
oldest open snapshot. This can lead to database size bloat for
high-update workloads when there are long-open snapshots and long-open
snapshot will be used for logical backup. By "almost new" I mean that the
key was updated more than once after the oldest snapshot.
If there were two snapshots with seq numbers s1 and s2 (s1 < s2), and if
we find two instances of the same key k1 that lie entirely within s1 and
s2 (i.e. s1 < k1 < s2), then the earlier version
of k1 can be safely deleted because that version is not visible in any snapshot.
Test Plan:
unit test attached
make clean check
Differential Revision: https://reviews.facebook.net/D6999
Summary:
Print out status at the end of a compaction run. This helps in
debugging.
Test Plan: make clean check
Reviewers: sheki
Reviewed By: sheki
Differential Revision: https://reviews.facebook.net/D7035
Summary:
This option is needed for fast bulk uploads. The goal is to load
all the data into files in L0 without any interference from
background compactions.
Test Plan: make clean check
Reviewers: sheki
Reviewed By: sheki
CC: leveldb
Differential Revision: https://reviews.facebook.net/D6849
Summary:
dbstress has an option to reopen the database. Make it such that the
previous handle is not closed before we reopen, this simulates a
situation similar to a process crash.
Added new api to DMImpl to remove the lock file.
Test Plan: run db_stress
Reviewers: emayanke
Reviewed By: emayanke
CC: leveldb
Differential Revision: https://reviews.facebook.net/D6777
Summary:
There are applications that operate on multiple leveldb instances.
These applications will like to pass in an opaque type for each
leveldb instance and this type should be passed back to the application
with every invocation of the CompactionFilter api.
Test Plan: Enehanced unit test for opaque parameter to CompactionFilter.
Reviewers: heyongqiang
Reviewed By: heyongqiang
CC: MarkCallaghan, sheki, emayanke
Differential Revision: https://reviews.facebook.net/D6711
Summary: Record BloomFliter hits and drop off reasons during compaction.
Test Plan: Unit tests work.
Reviewers: dhruba, heyongqiang
Reviewed By: dhruba
Differential Revision: https://reviews.facebook.net/D6591
Summary:
disable size compaction in ldb reduce_levels, this will avoid compactions rather than the manual comapction,
added --compression=none|snappy|zlib|bzip2 and --file_size= per-file size to ldb reduce_levels command
Test Plan: run ldb
Reviewers: dhruba, MarkCallaghan
Reviewed By: dhruba
CC: sheki, emayanke
Differential Revision: https://reviews.facebook.net/D6597
Summary:
When a new version is created, we sort all the files at every
level based on their size. This is necessary because we want
to compact the largest file first. The sorting takes quite a
bit of CPU.
Moved the sorting code to be outside the mutex. Also, the
earlier code was sorting files at all levels but we do not
need to sort the highest-number level because those files
are never the cause of any compaction. To reduce sorting
costs, we sort only the first few files in each level
because it is likely that those are the only files in that
level that will be picked for compaction.
At steady state, I have seen that this patch increase
throughout from 1500 writes/sec to 1700 writes/sec at the
end of a 72 hour run. The cpu saving by not sorting the
last level was not distinctive in this test run because
there were only 100K files in the highest numbered level.
I expect the cpu saving to be significant when the number of
files is much higher.
This is mostly an early preview and not ready for rigorous review.
With this patch, the writs/sec is now bottlenecked not by the sorting code but by GetOverlappingInputs. I am working on a patch to optimize GetOverlappingInputs.
Test Plan: make check
Reviewers: MarkCallaghan, heyongqiang
Reviewed By: heyongqiang
Differential Revision: https://reviews.facebook.net/D6411
Summary:
Added a conditional flush in ~DBImpl to flush.
There is still a chance of writes not being persisted if there is a
crash (not a clean shutdown) before the DBImpl instance is destroyed.
Test Plan: modified db_test to meet the new expectations.
Reviewers: dhruba, heyongqiang
Differential Revision: https://reviews.facebook.net/D6519
Summary:
The default compilation process now uses "-Wall" to compile.
Fix all compilation error generated by gcc.
Test Plan: make all check
Reviewers: heyongqiang, emayanke, sheki
Reviewed By: heyongqiang
CC: MarkCallaghan
Differential Revision: https://reviews.facebook.net/D6525
Summary:
There are certain use-cases where the application intends to
delete older keys aftre they have expired a certian time period.
One option for those applications is to periodically scan the
entire database and delete appropriate keys.
A better way is to allow the application to hook into the
compaction process. This patch allows the application to set
a method callback for every key that is being compacted. If
this method returns true, then the key is not preserved in
the output of the compaction.
Test Plan:
This is mostly to preview the proposed new public api.
Since it is a public api, please do due diligence on reviewing it.
I will be writing test cases for this api in mynext version of
this patch.
Reviewers: MarkCallaghan, heyongqiang
Reviewed By: heyongqiang
CC: sheki, adsharma
Differential Revision: https://reviews.facebook.net/D6285
Summary:
as subject
Test Plan:
run db_bench and db_test
Reviewers: dhruba
Reviewed By: dhruba
Differential Revision: https://reviews.facebook.net/D6111
Summary:
This makes the stall timers in MakeRoomForWrite more accurate by timing
the sleeps. From looking at the logs the real sleep times are usually
about 2000 usecs each when SleepForMicros(1000) is called. The modified LOG messages are:
2012/10/29-12:06:33.271984 2b3cc872f700 delaying write 13 usecs for level0_slowdown_writes_trigger
2012/10/29-12:06:34.688939 2b3cc872f700 delaying write 1728 usecs for rate limits with max score 3.83
Task ID: #
Blame Rev:
Test Plan:
run db_bench, look at DB/LOG
Revert Plan:
Database Impact:
Memcache Impact:
Other Notes:
EImportant:
- begin *PUBLIC* platform impact section -
Bugzilla: #
- end platform impact -
Reviewers: dhruba
Reviewed By: dhruba
Differential Revision: https://reviews.facebook.net/D6297
Summary:
The leveldb API is enhanced to support different compression algorithms at
different levels.
This adds the option min_level_to_compress to db_bench that specifies
the minimum level for which compression should be done when
compression is enabled. This can be used to disable compression for levels
0 and 1 which are likely to suffer from stalls because of the CPU load
for memtable flushes and (L0,L1) compaction. Level 0 is special as it
gets frequent memtable flushes. Level 1 is special as it frequently
gets all:all file compactions between it and level 0. But all other levels
could be the same. For any level N where N > 1, the rate of sequential
IO for that level should be the same. The last level is the
exception because it might not be full and because files from it are
not read to compact with the next larger level.
The same amount of time will be spent doing compaction at any
level N excluding N=0, 1 or the last level. By this standard all
of those levels should use the same compression. The difference is that
the loss (using more disk space) from a faster compression algorithm
is less significant for N=2 than for N=3. So we might be willing to
trade disk space for faster write rates with no compression
for L0 and L1, snappy for L2, zlib for L3. Using a faster compression
algorithm for the mid levels also allows us to reclaim some cpu
without trading off much loss in disk space overhead.
Also note that little is to be gained by compressing levels 0 and 1. For
a 4-level tree they account for 10% of the data. For a 5-level tree they
account for 1% of the data.
With compression enabled:
* memtable flush rate is ~18MB/second
* (L0,L1) compaction rate is ~30MB/second
With compression enabled but min_level_to_compress=2
* memtable flush rate is ~320MB/second
* (L0,L1) compaction rate is ~560MB/second
This practicaly takes the same code from https://reviews.facebook.net/D6225
but makes the leveldb api more general purpose with a few additional
lines of code.
Test Plan: make check
Differential Revision: https://reviews.facebook.net/D6261
Summary:
Adds the "MB/sec in" and "MB/sec out" to this line:
Amplification: 1.7 rate, 0.01 GB in, 0.02 GB out, 8.24 MB/sec in, 13.75 MB/sec out
Changes all values to be reported per interval and since test start for this line:
... thread 0: (10000,60000) ops and (19155.6,27307.5) ops/second in (0.522041,2.197198) seconds
Task ID: #
Blame Rev:
Test Plan:
run db_bench
Revert Plan:
Database Impact:
Memcache Impact:
Other Notes:
EImportant:
- begin *PUBLIC* platform impact section -
Bugzilla: #
- end platform impact -
Reviewers: dhruba
Reviewed By: dhruba
Differential Revision: https://reviews.facebook.net/D6291
Summary:
Adds a method that returns the score for the next level that most
needs compaction. That method is then used by db_bench to rate limit threads.
Threads are put to sleep at the end of each stats interval until the score
is less than the limit. The limit is set via the --rate_limit=$double option.
The specified value must be > 1.0. Also adds the option --stats_per_interval
to enable additional metrics reported every stats interval.
Task ID: #
Blame Rev:
Test Plan:
run db_bench
Revert Plan:
Database Impact:
Memcache Impact:
Other Notes:
EImportant:
- begin *PUBLIC* platform impact section -
Bugzilla: #
- end platform impact -
Reviewers: dhruba
Reviewed By: dhruba
Differential Revision: https://reviews.facebook.net/D6243
Summary: Enable LevelDb to create a new log file if current log file is too large.
Test Plan:
Write a script and manually check the generated info LOG.
Task ID: 1803577
Blame Rev:
Reviewers: dhruba, heyongqiang
Reviewed By: heyongqiang
CC: zshao
Differential Revision: https://reviews.facebook.net/D6003
Summary:
I used server uptime to compute per-level IO throughput rates. I
intended to use time spent doing compaction at that level. This fixes that.
Task ID: #
Blame Rev:
Test Plan:
run db_bench, look at results
Revert Plan:
Database Impact:
Memcache Impact:
Other Notes:
EImportant:
- begin *PUBLIC* platform impact section -
Bugzilla: #
- end platform impact -
Reviewers: dhruba
Reviewed By: dhruba
Differential Revision: https://reviews.facebook.net/D6237
Summary:
This adds more statistics to be reported by GetProperty("leveldb.stats").
The new stats include time spent waiting on stalls in MakeRoomForWrite.
This also includes the total amplification rate where that is:
(#bytes of sequential IO during compaction) / (#bytes from Put)
This also includes a lot more data for the per-level compaction report.
* Rn(MB) - MB read from level N during compaction between levels N and N+1
* Rnp1(MB) - MB read from level N+1 during compaction between levels N and N+1
* Wnew(MB) - new data written to the level during compaction
* Amplify - ( Write(MB) + Rnp1(MB) ) / Rn(MB)
* Rn - files read from level N during compaction between levels N and N+1
* Rnp1 - files read from level N+1 during compaction between levels N and N+1
* Wnp1 - files written to level N+1 during compaction between levels N and N+1
* NewW - new files written to level N+1 during compaction
* Count - number of compactions done for this level
This is the new output from DB::GetProperty("leveldb.stats"). The old output stopped at Write(MB)
Compactions
Level Files Size(MB) Time(sec) Read(MB) Write(MB) Rn(MB) Rnp1(MB) Wnew(MB) Amplify Read(MB/s) Write(MB/s) Rn Rnp1 Wnp1 NewW Count
-------------------------------------------------------------------------------------------------------------------------------------
0 3 6 33 0 576 0 0 576 -1.0 0.0 1.3 0 0 0 0 290
1 127 242 351 5316 5314 570 4747 567 17.0 12.1 12.1 287 2399 2685 286 32
2 161 328 54 822 824 326 496 328 4.0 1.9 1.9 160 251 411 160 161
Amplification: 22.3 rate, 0.56 GB in, 12.55 GB out
Uptime(secs): 439.8
Stalls(secs): 206.938 level0_slowdown, 0.000 level0_numfiles, 24.129 memtable_compaction
Task ID: #
Blame Rev:
Test Plan:
run db_bench
Revert Plan:
Database Impact:
Memcache Impact:
Other Notes:
EImportant:
- begin *PUBLIC* platform impact section -
Bugzilla: #
- end platform impact -
(cherry picked from commit ecdeead38f86cc02e754d0032600742c4f02fec8)
Reviewers: dhruba
Differential Revision: https://reviews.facebook.net/D6153
Summary:
The compaction process deletes a large number of files. This takes
quite a bit of time and is best done outside the mutex lock.
Test Plan: make check
Differential Revision: https://reviews.facebook.net/D6123
Summary:
The compaction process deletes a large number of files. This takes
quite a bit of time and is best done outside the mutex lock.
Test Plan: make check
Differential Revision: https://reviews.facebook.net/D6123
Summary:
The compaction process deletes a large number of files. This takes
quite a bit of time and is best done outside the mutex lock.
Test Plan:
Reviewers:
CC:
Task ID: #
Blame Rev:
published in https://reviews.facebook.net/D5997.
Summary:
This patch allows compaction to occur in multiple background threads
concurrently.
If a manual compaction is issued, the system falls back to a
single-compaction-thread model. This is done to ensure correctess
and simplicity of code. When the manual compaction is finished,
the system resumes its concurrent-compaction mode automatically.
The updates to the manifest are done via group-commit approach.
Test Plan: run db_bench
Summary:
The method DeleteObsolete files is a very costly methind, especially
when the number of files in a system is large. It makes a list of
all live-files and then scans the directory to compute the diff.
By default, this method is executed after every compaction run.
This patch makes it such that DeleteObsolete files is never
invoked twice within a configured period.
Test Plan: run all unit tests
Reviewers: heyongqiang, MarkCallaghan
Reviewed By: MarkCallaghan
Differential Revision: https://reviews.facebook.net/D6045
Summary:
If ReadCompaction is switched off, then it is better to not even
submit background compaction jobs. I see about 3% increase in
read-throughput on a pure memory database.
Test Plan: run db_bench
Reviewers: heyongqiang
Reviewed By: heyongqiang
Differential Revision: https://reviews.facebook.net/D5673
Summary: Print out the compile version in the LOG.
Test Plan: run dbbench and verify LOG
Reviewers: heyongqiang
Reviewed By: heyongqiang
Differential Revision: https://reviews.facebook.net/D5529
Summary:
A set of apis that allows an application to backup data from the
leveldb database based on a set of files.
Test Plan: unint test attached. more coming soon.
Reviewers: heyongqiang
Reviewed By: heyongqiang
Differential Revision: https://reviews.facebook.net/D5439
Summary: added a new option db_log_dir, which points the log dir. Inside that dir, in order to make log names unique, the log file name is prefixed with the leveldb data dir absolute path.
Test Plan: db_test
Reviewers: dhruba
Reviewed By: dhruba
Differential Revision: https://reviews.facebook.net/D5205
Summary:
Clean up compiler warnings generated by -Wall option.
make clean all OPT=-Wall
This is a pre-requisite before making a new release.
Test Plan: compile and run unit tests
Reviewers: heyongqiang
Reviewed By: heyongqiang
Differential Revision: https://reviews.facebook.net/D5019
Summary: as subject. ported the change from google code leveldb 1.5
Test Plan: run db_test
Reviewers: dhruba
Differential Revision: https://reviews.facebook.net/D4839
Summary:
Introduce a new method Env->Fsync() that issues fsync (instead of fdatasync).
This is needed for data durability when running on ext3 filesystems.
Added options to the benchmark db_bench to generate performance numbers
with either fsync or fdatasync enabled.
Cleaned up Makefile to build leveldb_shell only when building the thrift
leveldb server.
Test Plan: build and run benchmark
Reviewers: heyongqiang
Reviewed By: heyongqiang
Differential Revision: https://reviews.facebook.net/D4911
Summary: Log the open-options to the LOG. Use options_ instead of options because SanitizeOptions could modify the max_file_open limit.
Test Plan: num db_bench
Reviewers: heyongqiang
Reviewed By: heyongqiang
Differential Revision: https://reviews.facebook.net/D4833
Summary:
Fixed unit test c_test by initializing logger=NULL.
Removed "atomic" from last_log_ts so that unit tests do not require C11 compiler.
Anyway, last_log_ts is mostly used for logging, so it is ok if it is loosely
accurate.
Test Plan: run c_test
Reviewers: heyongqiang
Reviewed By: heyongqiang
Differential Revision: https://reviews.facebook.net/D4803
Summary: Record the version of the source that we are compiling. We keep a record of the git revision in util/version.cc. This source file is then built as a regular source file as part of the compilation process. One can run "strings executable_filename | grep _build_" to find the version of the source that we used to build the executable file.
Test Plan: none
Differential Revision: https://reviews.facebook.net/D4785
Summary:
as subject.
A new log is written to scribe via thrift client when a new db is opened and when there is
a compaction.
a new option var scribe_log_db_stats is added.
Test Plan: manually checked using command "ptail -time 0 leveldb_deploy_stats"
Reviewers: dhruba
Differential Revision: https://reviews.facebook.net/D4659
Summary: as subject and only maintain 10 log files.
Test Plan: new test in db_test
Reviewers: dhruba
Differential Revision: https://reviews.facebook.net/D4731
Summary: as subject. The flush will flush everything in the db.
Test Plan: new test in db_test.cc
Reviewers: dhruba
Reviewed By: dhruba
Differential Revision: https://reviews.facebook.net/D4029
Summary:
Added option --writes=xxx to specify the number of keys that we want to overwrite in the benchmark.
Task ID: #
Blame Rev:
Test Plan: Revert Plan:
Reviewers: adsharma
CC: sc
Differential Revision: https://reviews.facebook.net/D3465
In particular, we add a new FilterPolicy class. An instance
of this class can be supplied in Options when opening a
database. If supplied, the instance is used to generate
summaries of keys (e.g., a bloom filter) which are placed in
sstables. These summaries are consulted by DB::Get() so we
can avoid reading sstable blocks that are guaranteed to not
contain the key we are looking for.
This change provides one implementation of FilterPolicy
based on bloom filters.
Other changes:
- Updated version number to 1.4.
- Some build tweaks.
- C binding for CompactRange.
- A few more benchmarks: deleteseq, deleterandom, readmissing, seekrandom.
- Minor .gitignore update.
- Replace raw slice comparison with a call to user comparator.
Added test for custom comparators.
- Fix end of namespace comments.
- Fixed bug in picking inputs for a level-0 compaction.
When finding overlapping files, the covered range may expand
as files are added to the input set. We now correctly expand
the range when this happens instead of continuing to use the
old range. For example, suppose L0 contains files with the
following ranges:
F1: a .. d
F2: c .. g
F3: f .. j
and the initial compaction target is F3. We used to search
for range f..j which yielded {F2,F3}. However we now expand
the range as soon as another file is added. In this case,
when F2 is added, we expand the range to c..j and restart the
search. That picks up file F1 as well.
This change fixes a bug related to deleted keys showing up
incorrectly after a compaction as described in Issue 44.
(Sync with upstream @25072954)
- Added DB::CompactRange() method.
Changed manual compaction code so it breaks up compactions of
big ranges into smaller compactions.
Changed the code that pushes the output of memtable compactions
to higher levels to obey the grandparent constraint: i.e., we
must never have a single file in level L that overlaps too
much data in level L+1 (to avoid very expensive L-1 compactions).
Added code to pretty-print internal keys.
- Fixed bug where we would not detect overlap with files in
level-0 because we were incorrectly using binary search
on an array of files with overlapping ranges.
Added "leveldb.sstables" property that can be used to dump
all of the sstables and ranges that make up the db state.
- Removing post_write_snapshot support. Email to leveldb mailing
list brought up no users, just confusion from one person about
what it meant.
- Fixing static_cast char to unsigned on BIG_ENDIAN platforms.
Fixes Issue 35 and Issue 36.
- Comment clarification to address leveldb Issue 37.
- Change license in posix_logger.h to match other files.
- A build problem where uint32 was used instead of uint32_t.
Sync with upstream @24408625
- Fix bug in Get: when it triggers a compaction, it could sometimes
mark the compaction with the wrong level (if there was a gap
in the set of levels examined for the Get).
- Do not hold mutex while writing to the log file or to the
MANIFEST file.
Added a new benchmark that runs a writer thread concurrently with
reader threads.
Percentiles
------------------------------
micros/op: avg median 99 99.9 99.99 99.999 max
------------------------------------------------------
before: 42 38 110 225 32000 42000 48000
after: 24 20 55 65 130 1100 7000
- Fixed race in optimized Get. It should have been using the
pinned memtables, not the current memtables.
git-svn-id: https://leveldb.googlecode.com/svn/trunk@50 62dab493-f737-651d-591e-8d6aee1b9529
- Fix for issue 33 (non-null-terminated result from
leveldb_property_value())
- Support for running multiple instances of a benchmark in parallel.
- Reduce lock contention on Get():
(1) Do not hold the lock while searching memtables.
(2) Shard block and table caches 16-ways.
Benchmark for evaluating this change:
$ db_bench --benchmarks=fillseq1,readrandom --threads=$n
(fillseq1 is a small hack to make sure fillseq runs once regardless
of number of threads specified on the command line).
git-svn-id: https://leveldb.googlecode.com/svn/trunk@49 62dab493-f737-651d-591e-8d6aee1b9529
- Removed one copy of an uncompressed block contents changing
the signature of Snappy_Uncompress() so it uncompresses into a
flat array instead of a std::string.
Speeds up readrandom ~10%.
- Instead of a combination of Env/WritableFile, we now have a
Logger interface that can be easily overridden applications
that want to supply their own logging.
- Separated out the gcc and Sun Studio parts of atomic_pointer.h
so we can use 'asm', 'volatile' keywords for Sun Studio.
git-svn-id: https://leveldb.googlecode.com/svn/trunk@39 62dab493-f737-651d-591e-8d6aee1b9529
Slight tweak to the no-overlap optimization: only push to
level 2 to reduce the amount of wasted space when the same
small key range is being repeatedly overwritten.
Fix for Issue 18: Avoid failure on Windows by avoiding
deletion of lock file until the end of DestroyDB().
Fix for Issue 19: Disregard sequence numbers when checking for
overlap in sstable ranges. This fixes issue 19: when writing
the same key over and over again, we would generate a sequence
of sstables that were never merged together since their sequence
numbers were disjoint.
Don't ignore map/unmap error checks.
Miscellaneous fixes for small problems Sanjay found while diagnosing
issue/9 and issue/16 (corruption_testr failures).
- log::Reader reports the record type when it finds an unexpected type.
- log::Reader no longer reports an error when it encounters an expected
zero record regardless of the setting of the "checksum" flag.
- Added a missing forward declaration.
- Documented a side-effects of larger write buffer sizes
(longer recovery time).
git-svn-id: https://leveldb.googlecode.com/svn/trunk@37 62dab493-f737-651d-591e-8d6aee1b9529
- Implemented Get() directly instead of building on top of a full
merging iterator stack. This speeds up the "readrandom" benchmark
by up to 15-30%.
- Fixed an opensource compilation problem.
Added --db=<name> flag to control where the database is placed.
- Automatically compact a file when we have done enough
overlapping seeks to that file.
- Fixed a performance bug where we would read from at least one
file in a level even if none of the files overlapped the key
being read.
- Makefile fix for Mac OSX installations that have XCode 4 without XCode 3.
- Unified the two occurrences of binary search in a file-list
into one routine.
- Found and fixed a bug where we would unnecessarily search the
last file when looking for a key larger than all data in the
level.
- A fix to avoid the need for trivial move compactions and
therefore gets rid of two out of five syncs in "fillseq".
- Removed the MANIFEST file write when switching to a new
memtable/log-file for a 10-20% improvement on fill speed on ext4.
- Adding a SNAPPY setting in the Makefile for folks who have
Snappy installed. Snappy compresses values and speeds up writes.
git-svn-id: https://leveldb.googlecode.com/svn/trunk@32 62dab493-f737-651d-591e-8d6aee1b9529
Fixed race condition reported by Dave Smit (dizzyd@dizzyd,com)
on the leveldb mailing list. We were not signalling
waiters after a trivial move from level-0. The result was
that in some cases (hard to reproduce), a write would get
stuck forever waiting for the number of level-0 files to drop
below its hard limit.
The new code is simpler: there is just one condition variable
instead of two, and the condition variable is signalled after
every piece of background work finishes. Also, all compaction
work (including for manual compactions) is done in the
background thread, and therefore we can remove the
"compacting_" variable.
git-svn-id: https://leveldb.googlecode.com/svn/trunk@31 62dab493-f737-651d-591e-8d6aee1b9529
* Patch LevelDB to build for OSX and iOS
* Fix race condition in memtable iterator deletion.
* Other small fixes.
git-svn-id: https://leveldb.googlecode.com/svn/trunk@29 62dab493-f737-651d-591e-8d6aee1b9529
* env_chromium.cc should not export symbols.
* Fix MSVC warnings.
* Removed large value support.
* Fix broken reference to documentation file
git-svn-id: https://leveldb.googlecode.com/svn/trunk@24 62dab493-f737-651d-591e-8d6aee1b9529
