Summary:
The test was not using separate MemTablePostProcessInfo per memetable insert thread and thus tsan was complaining about data race.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5308
Differential Revision: D15356420
Pulled By: maysamyabandeh
fbshipit-source-id: 46c2f2d19fb02c3c775b587aa09ca9c0dae6ed04
Summary:
This PR has two fixes for crash test failures -
1. Fix a bug in TestMultiGet() in db_stress that was passing list of key to MultiGet() in the wrong order, thus ensuring that actual values don't match expected values
2. Remove an incorrect assertion in FilePickerMultiGet::GetNextFileInLevelWithKeys() that checks that files in a level are in sorted order. This is not true with MultiGet(), especially if there are duplicate keys and we may have to go back one file for the next key. Furthermore, this assertion makes more sense when a new version is created, rather than at lookup time
Test -
asan_crash and ubsan_crash tests
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5301
Differential Revision: D15337383
Pulled By: anand1976
fbshipit-source-id: 35092cb15bbc1700e5e823cbe07bfa62f1e9e6c6
Summary:
Performing unordered writes in rocksdb when unordered_write option is set to true. When enabled the writes to memtable are done without joining any write thread. This offers much higher write throughput since the upcoming writes would not have to wait for the slowest memtable write to finish. The tradeoff is that the writes visible to a snapshot might change over time. If the application cannot tolerate that, it should implement its own mechanisms to work around that. Using TransactionDB with WRITE_PREPARED write policy is one way to achieve that. Doing so increases the max throughput by 2.2x without however compromising the snapshot guarantees.
The patch is prepared based on an original by siying
Existing unit tests are extended to include unordered_write option.
Benchmark Results:
```
TEST_TMPDIR=/dev/shm/ ./db_bench_unordered --benchmarks=fillrandom --threads=32 --num=10000000 -max_write_buffer_number=16 --max_background_jobs=64 --batch_size=8 --writes=3000000 -level0_file_num_compaction_trigger=99999 --level0_slowdown_writes_trigger=99999 --level0_stop_writes_trigger=99999 -enable_pipelined_write=false -disable_auto_compactions --unordered_write=1
```
With WAL
- Vanilla RocksDB: 78.6 MB/s
- WRITER_PREPARED with unordered_write: 177.8 MB/s (2.2x)
- unordered_write: 368.9 MB/s (4.7x with relaxed snapshot guarantees)
Without WAL
- Vanilla RocksDB: 111.3 MB/s
- WRITER_PREPARED with unordered_write: 259.3 MB/s MB/s (2.3x)
- unordered_write: 645.6 MB/s (5.8x with relaxed snapshot guarantees)
- WRITER_PREPARED with unordered_write disable concurrency control: 185.3 MB/s MB/s (2.35x)
Limitations:
- The feature is not yet extended to `max_successive_merges` > 0. The feature is also incompatible with `enable_pipelined_write` = true as well as with `allow_concurrent_memtable_write` = false.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5218
Differential Revision: D15219029
Pulled By: maysamyabandeh
fbshipit-source-id: 38f2abc4af8780148c6128acdba2b3227bc81759
Summary:
Previous code may call `~ColumnFamilyData` in `DBImpl::AtomicFlushMemTablesToOutputFiles` if the column family is dropped or `cfd->IsFlushPending() == false`. In `~ColumnFamilyData`, the db mutex is released briefly and re-acquired. This can cause correctness issue. The reason is as follows.
Assume there are more bg flush threads. After bg_flush_thr1 releases the db mutex, bg_flush_thr2 can grab it and pop an element from the flush queue. This will cause bg_flush_thr2 to accidentally pick some memtables which should have been picked by bg_flush_thr1. To make the matter worse, bg_flush_thr2 can clear `flush_requested_` flag for the memtable list, causing a subsequent call to `MemTableList::IsFlushPending()` by bg_flush_thr1 to return false, which is wrong.
The fix is to delay `ColumnFamilyData::Unref` and `~ColumnFamilyData` for column families not selected for flush until `AtomicFlushMemTablesToOutputFiles` returns. Furthermore, a bg flush thread should not clear `MemTableList::flush_requested_` in `MemTableList::PickMemtablesToFlush` unless atomic flush is not used **or** the memtable list does not have unpicked memtables.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5294
Differential Revision: D15295297
Pulled By: riversand963
fbshipit-source-id: 03b101205ca22c242647cbf488bcf0ed80b2ecbd
Summary:
There were no C bindings for lowering thread pool priority. This adds those.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5285
Differential Revision: D15290050
Pulled By: siying
fbshipit-source-id: b2ed94d0c39d27434ace2204829a242b53d0d67a
Summary:
When reseek happens in merging iterator, reseeking a child iterator can be avoided if:
(1) the iterator represents imutable data
(2) reseek() to a larger key than the current key
(3) the current key of the child iterator is larger than the seek key
because it is guaranteed that the result will fall into the same position.
This optimization will be useful for use cases where users keep seeking to keys nearby in ascending order.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5286
Differential Revision: D15283635
Pulled By: siying
fbshipit-source-id: 35f79ffd5ce3609146faa8cd55f2bfd733502f83
Summary:
This PR fixes a couple of bugs in FilePickerMultiGet that were causing db_stress test failures. The failures were caused by -
1. Improper handling of a key that matches the user key portion of an L0 file's largest key. In this case, the curr_index_in_curr_level file index in L0 for that key was getting incremented, but batch_iter_ was not advanced. By design, all keys in a batch are supposed to be checked against an L0 file before advancing to the next L0 file. Not advancing to the next key in the batch was causing a double increment of curr_index_in_curr_level due to the same key being processed again
2. Improper handling of a key that matches the user key portion of the largest key in the last file of L1 and higher. This was resulting in a premature end to the processing of the batch for that level when the next key in the batch is a duplicate. Typically, the keys in MultiGet will not be duplicates, but its good to handle that case correctly
Test -
asan_crash
make check
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5292
Differential Revision: D15282530
Pulled By: anand1976
fbshipit-source-id: d1a6a86e0af273169c3632db22a44d79c66a581f
Summary:
Right now, DBIter::Next() always checks whether an entry is for the same user key as the previous entry to see whether the key should be hidden to the user. However, if previous entry's sequence number is 0, the check is not needed because 0 is the oldest possible sequence number.
We could extend it from seqnum 0 case to simply prev_seqno >= current_seqno. However, it is less robust with bug or unexpected situations, while the gain is relatively low. We can always extend it later when needed.
In a readseq benchmark with full formed LSM-tree, number of key comparisons called is reduced from 2.981 to 2.165. readseq against a fully compacted DB, no key comparison is called. Performance in this benchmark didn't show obvious improvement, which is expected because key comparisons only takes small percentage of CPU. But it may show up to be more effective if users have an expensive customized comparator.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5244
Differential Revision: D15067257
Pulled By: siying
fbshipit-source-id: b7e1ef3ec4fa928cba509683d2b3246e35d270d9
Summary:
Previously in PR https://github.com/facebook/rocksdb/pull/5161 we have added the capability to do WAL tailing in `OpenAsSecondary`, in this PR we extend such feature to `TryCatchUpWithPrimary` which is useful for an secondary RocksDB instance to retrieve and apply the latest updates and refresh log readers if needed.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5282
Differential Revision: D15261011
Pulled By: miasantreble
fbshipit-source-id: a15c94471e8c3b3b1f7f47c3135db1126e936949
Summary:
Part of compaction cpu goes to processing snapshot list, the larger the list the bigger the overhead. Although the lifetime of most of the snapshots is much shorter than the lifetime of compactions, the compaction conservatively operates on the list of snapshots that it initially obtained. This patch allows the snapshot list to be updated via a callback if the compaction is taking long. This should let the compaction to continue more efficiently with much smaller snapshot list.
For simplicity, to avoid the feature is disabled in two cases: i) When more than one sub-compaction are sharing the same snapshot list, ii) when Range Delete is used in which the range delete aggregator has its own copy of snapshot list.
This fixes the reverted https://github.com/facebook/rocksdb/pull/5099 issue with range deletes.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5278
Differential Revision: D15203291
Pulled By: maysamyabandeh
fbshipit-source-id: fa645611e606aa222c7ce53176dc5bb6f259c258
Summary:
This PR fixes three memory issues found by ASAN
* in db_stress, the key vector for MultiGet is created using `emplace_back` which could potentially invalidates references to the underlying storage (vector<string>) due to auto resizing. Fix by calling reserve in advance.
* Similar issue in construction of GetContext autovector in version_set.cc
* In multiget_context.h use T[] specialization for unique_ptr that holds a char array
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5279
Differential Revision: D15202893
Pulled By: miasantreble
fbshipit-source-id: 14cc2cda0ed64d29f2a1e264a6bfdaa4294ee75d
Summary:
Right now, when Seek() is called again, RocksDB always does a binary search against the files and index blocks, even if they end up with the same file/block. Improve it as following:
1. in LevelIterator, reseek first try to check the boundary of the current file. If it falls into the same file, skip the binary search to find the file
2. in block based table iterator, reseek skip to reseek the iterator block if the seek key is larger than the current key and lower than the index key (boundary of the current block and the next block).
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5256
Differential Revision: D15105072
Pulled By: siying
fbshipit-source-id: 39634bdb4a881082451fa39cecd7ecf12160bf80
Summary:
Sometimes, users might make mistake of not releasing snapshots before closing the DB. This is undocumented use of RocksDB and the behavior is unknown. We return DB::Close() to provide a way to check it for the users. Aborted() will be returned to users when they call DB::Close().
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5272
Differential Revision: D15159713
Pulled By: siying
fbshipit-source-id: 39369def612398d9f239d83d396b5a28e5af65cd
Summary:
Our daily stress tests are failing after this feature. Reverting temporarily until we figure the reason for test failures.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5269
Differential Revision: D15151285
Pulled By: maysamyabandeh
fbshipit-source-id: e4002b99690a97df30d4b4b58bf0f61e9591bc6e
Summary:
With atomic flush, RocksDB background flush will flush memtables of a column family up to the largest memtable id in the immutable memtable list. This can introduce a bug in the following scenario. A user thread inserts into a column family until the memtable is full and triggers a flush. This will add the column family to flush_scheduler_. Then the user thread writes another record to the column family. In the PreprocessWrite function, the user thread picks the column family from flush_scheduler_ and schedules a flush request. The flush request gaurantees to flush all the memtables up to the current largest memtable ID of the immutable memtable list. Then the user thread writes new data to the newly-created active memtable. After the write returns, the user thread closes the db. This can cause assertion failure when the background flush thread tries to install superversion for the column family. The solution is to not install flush results if the db has already set `shutting_down_` to true.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5254
Differential Revision: D15124149
Pulled By: riversand963
fbshipit-source-id: 0a667a41339dedb5a18bcb01b0bf11c275c04df0
Summary:
Improve the iterators performance when the user explicitly sets the readahead size via `ReadOptions.readahead_size`.
1. Stop creating new table readers when the user explicitly sets readahead size.
2. Make use of an internal buffer based on `FilePrefetchBuffer` instead of using `ReadaheadRandomAccessFileReader`, to handle the user readahead requests (for both buffered and direct io cases).
3. Add `readahead_size` to db_bench.
**Benchmarks:**
https://gist.github.com/sagar0/53693edc320a18abeaeca94ca32f5737
For 1 MB readahead, Buffered IO performance improves by 28% and Direct IO performance improves by 50%.
For 512KB readahead, Buffered IO performance improves by 30% and Direct IO performance improves by 67%.
**Test Plan:**
Updated `DBIteratorTest.ReadAhead` test to make sure that:
- no new table readers are created for iterators on setting ReadOptions.readahead_size
- At least "readahead" number of bytes are actually getting read on each iterator read.
TODO later:
- Use similar logic for compactions as well.
- This ties in nicely with #4052 and paves the way for removing ReadaheadRandomAcessFile later.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5246
Differential Revision: D15107946
Pulled By: sagar0
fbshipit-source-id: 2c1149729ca7d779e4e8b7710ba6f4e8cbfd3bea
Summary:
The newly added test CompactionJobTest.SnapshotRefresh sets the snapshot refresh period to 0 to stress the feature. This results into large number of refresh events, which in turn results into an UBSAN failure when a bitwise shift operand goes beyond the uint64_t size.
The patch fixes that by simplifying the shift logic to be done only by 2 bits after each refresh. Furthermore it verifies that the shift operation does not result in decreasing the refresh period.
Testing:
COMPILE_WITH_UBSAN=1 make -j32 compaction_job_test
./compaction_job_test --gtest_filter=CompactionJobTest.SnapshotRefresh
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5257
Differential Revision: D15106463
Pulled By: maysamyabandeh
fbshipit-source-id: f2718898ea7ba4fa9f7e87b70cf98fe647c0de80
Summary:
Part of compaction cpu goes to processing snapshot list, the larger the list the bigger the overhead. Although the lifetime of most of the snapshots is much shorter than the lifetime of compactions, the compaction conservatively operates on the list of snapshots that it initially obtained. This patch allows the snapshot list to be updated via a callback if the compaction is taking long. This should let the compaction to continue more efficiently with much smaller snapshot list.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5099
Differential Revision: D15086710
Pulled By: maysamyabandeh
fbshipit-source-id: 7649f56c3b6b2fb334962048150142a3bf9c1a12
Summary:
Currently one thread in RocksDB keeps a WAL file open while another thread
deletes it. Although the first thread never writes to the WAL again, it still
tries to close it in the end. This is fine on POSIX, but can be problematic on
other platforms, e.g. HDFS, etc.. It will either cause a lot of warning messages or
throw exceptions. The solution is to let the second thread close the WAL before deleting it.
RocksDB keeps the writers of the logs to delete in `logs_to_free_`, which is passed to `job_context` during `FindObsoleteFiles` (holding mutex). Then in `PurgeObsoleteFiles` (without mutex), these writers should close the logs.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5233
Differential Revision: D15032670
Pulled By: riversand963
fbshipit-source-id: c55e8a612db8cc2306644001a5e6d53842a8f754
Summary:
We have a DB with ~4k column families and ~70k files. On shutdown, destroying the 4k ColumnFamilyHandle-s takes over 2 minutes. Most of this time is spent in VersionSet::AddLiveFiles() called from FindObsoleteFiles() from ~ColumnFamilyHandleImpl(). It's just iterating over the list of files in memory. This seems completely unnecessary as no obsolete files are actually found since the CFs are not even dropped. This PR fixes that.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5238
Differential Revision: D15056342
Pulled By: siying
fbshipit-source-id: 2aa342ef3770b4aa384ce81f8768e485480e4f08
Summary: PR https://github.com/facebook/rocksdb/pull/4899 implemented the general framework for RocksDB secondary instances. This PR adds the support for WAL tailing in `OpenAsSecondary`, which means after the `OpenAsSecondary` call, the secondary is now able to see primary's writes that are yet to be flushed. The secondary can see primary's writes in the WAL up to the moment of `OpenAsSecondary` call starts.
Differential Revision: D15059905
Pulled By: miasantreble
fbshipit-source-id: 44f71f548a30b38179a7940165e138f622de1f10
Summary:
In some cases, we want to known the smallest and largest sequence numbers of sstable files, to help us get more details.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5231
Differential Revision: D15038087
Pulled By: siying
fbshipit-source-id: c473c1ca07b53efe2f1884fa1ecdc8686f455ed8
Summary:
It's hard to get DBIter to directly use InternalIterator::NextAndGetResult() because the code change would be complicated. Instead, use IteratorWrapper, where Next() is already using NextAndGetResult(). Performance number is hard to measure because it is small and ther is variation. I run readseq many times, and there seems to be 1% gain.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5214
Differential Revision: D15003635
Pulled By: siying
fbshipit-source-id: 17af1965c409c2fe90cd85037fbd2c5a1364f82a
Summary:
Introduce BlockBasedTableOptions::index_shortening to give users control on which key shortening techniques to be used in building index blocks. Before this patch, both separators and successor keys where shortened in indexes. With this patch, the default is set to kShortenSeparators to only shorten the separators. Since each index block has many separators and only one successor (last key), the change should not have negative impact on index block size. However it should prevent many unnecessary block loads where due to approximation introduced by shorted successor, seek would land us to the previous block and then fix it by moving to the next one.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5174
Differential Revision: D14884185
Pulled By: al13n321
fbshipit-source-id: 1b08bc8c03edcf09b6b8c16e9a7eea08ad4dd534
Summary:
Savepoints are assumed to be used in a stack-wise fashion (only
the top element should be used), so they were stored by `WriteBatch`
in a member variable `save_points` using an std::stack.
Conceptually this is fine, but the implementation had a few issues:
- the `save_points_` instance variable was a plain pointer to a heap-
allocated `SavePoints` struct. The destructor of `WriteBatch` simply
deletes this pointer. However, the copy constructor of WriteBatch
just copied that pointer, meaning that copying a WriteBatch with
active savepoints will very likely have crashed before. Now a proper
copy of the savepoints is made in the copy constructor, and not just
a copy of the pointer
- `save_points_` was an std::stack, which defaults to `std::deque` for
the underlying container. A deque is a bit over the top here, as we
only need access to the most recent savepoint (i.e. stack.top()) but
never any elements at the front. std::deque is rather expensive to
initialize in common environments. For example, the STL implementation
shipped with GNU g++ will perform a heap allocation of more than 500
bytes to create an empty deque object. Although the `save_points_`
container is created lazily by RocksDB, moving from a deque to a plain
`std::vector` is much more memory-efficient. So `save_points_` is now
a vector.
- `save_points_` was changed from a plain pointer to an `std::unique_ptr`,
making ownership more explicit.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5192
Differential Revision: D15024074
Pulled By: maysamyabandeh
fbshipit-source-id: 5b128786d3789cde94e46465c9e91badd07a25d7
Summary:
My compiler doesn't inline DBIter::Next() to arena wrapped iterator, even if it is a direct forward. Adding this annotation makes it inlined. It might not always work but inlinging this function to arena wrapped iterator always feels like the right decision.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5217
Differential Revision: D15004086
Pulled By: siying
fbshipit-source-id: a4cffd79c6fb092669a3a90633c9aa5e494f8a66
Summary:
We found an issue in Periodic Compactions (introduced in #5166) where files were not being picked up for compactions as all the SST files created with older versions of RocksDB have `file_creation_time` as 0. (Note that `file_creation_time` is a new table property introduced in #5166).
To address this, Periodic compactions now fall back to looking at the `creation_time` table property or the file's modification time (as given by the Env) when `file_creation_time` table property is found to be 0.
Here how the file's modification time (and, in turn, the file age) is computed now:
1. Use `file_creation_time` table property if it is > 0.
1. If not, then use `creation_time` table property if it is > 0.
1. If not, then use file's mtime stat metadata given by the underlying Env.
Don't consider the file at all for compaction if the modification time cannot be correctly determined based on the above conditions.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5184
Differential Revision: D14907795
Pulled By: sagar0
fbshipit-source-id: 4bb2f3631f9a3e04470c674a1d13544584e1e56c
Summary:
Reorganize the code so that no function call into ReadRangeDelAggregator is needed if there is no tomb range stone.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5202
Differential Revision: D14968155
Pulled By: siying
fbshipit-source-id: 0bd61911293c7a27b4e1b8d57c66d0c4ad6a6a5f
Summary:
Several small changes for Next():
1. Reducing branching by always update local_stats_.next_count_++ even if statistics is null. This should be faster than a branching.
2. Replacing ResetInternalKeysSkippedCounter() in Next() because the valid_ check is not needed in this case.
3. iter_->Valid() should always be true for non merge case. Remove this check.
4. Adding an inline annotation. It ends up with not picked up by my compiler, but it shouldn't hurt.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5200
Differential Revision: D15000391
Pulled By: siying
fbshipit-source-id: be97f61c708968234fb8e5cf272b5c2ac07dc4dd
Summary:
In long scans, virtual function calls of Next(), Valid(), key() and value() are not trivial. By introducing NextAndGetResult(), Some of the Next(), Valid() and key() calls are consolidated into one virtual function call to reduce CPU.
Also did some inline tricks and add some "final" randomly in some functions. Even without the "final" annotation, most Next() calls are inlined with -O3, but sometimes with a final it is inlined by O2 too. It doesn't hurt to add those final annotations.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5197
Differential Revision: D14945977
Pulled By: siying
fbshipit-source-id: 7003969f9a5f1d5717f0bda503b91d19ba75ed88
Summary:
`GetOverlappingInputsRangeBinarySearch` firstly use binary search
to find a index in the given range `[begin, end]`. But after find
the index, then use linear search to find the `start_index` and
`end_index`. So the search process degraded to linear time.
Here optmize the search process with below changes:
- use `std::lower_bound` and `std::upper_bound` to get
`lg(n)` search complexity.
- use uniformed lambda for search process.
- simplify process for `within_interval` true or false.
- remove function `ExtendFileRangeWithinInterval`
and `ExtendFileRangeOverlappingInterval`.
Signed-off-by: JiYou <jiyou09@gmail.com>
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4987
Differential Revision: D14984192
Pulled By: riversand963
fbshipit-source-id: fae4b8e59a21b7e350718d60cdc94dd55ac81e89
Summary:
this PR fixes the following compile warning:
```
db/memtable.cc: In member function ‘virtual void rocksdb::MemTableIterator::Seek(const rocksdb::Slice&)’:
db/memtable.cc:321:22: error: declaration of ‘user_key’ shadows a member of 'this' [-Werror=shadow]
Slice user_key(ExtractUserKey(k));
^
db/memtable.cc: In member function ‘virtual void rocksdb::MemTableIterator::SeekForPrev(const rocksdb::Slice&)’:
db/memtable.cc:338:22: error: declaration of ‘user_key’ shadows a member of 'this' [-Werror=shadow]
Slice user_key(ExtractUserKey(k));
^
```
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5204
Differential Revision: D14970160
Pulled By: miasantreble
fbshipit-source-id: 388eb089f90c4528cc6d615dd4607fb53ceac705
Summary:
Depending on the config, manual compaction (leveled compaction style) does following compactions:
L0->L1
L1->L2
...
Ln-1 -> Ln
Ln -> Ln
The final Ln -> Ln compaction is partly unnecessary as it recompacts all the files that were just generated by the Ln-1 -> Ln. We should avoid recompacting such files. This rule should be applied to Lmax only.
Resolves issue https://github.com/facebook/rocksdb/issues/4995
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5138
Differential Revision: D14940106
Pulled By: miasantreble
fbshipit-source-id: 8d3cf5507a17e76f3333cfd4bac5256d005636e5
Summary:
Dummy cache size of 1MB is too large for small block sizes. Our GetDefaultCacheShardBits() use min_shard_size = 512L * 1024L to determine number of shards, so 1MB will excceeds the size of the whole shard and make the cache excceeds the budget.
Change it to 256KB accordingly.
There shouldn't be obvious performance impact, since inserting a cache entry every 256KB of memtable inserts is still infrequently enough.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5175
Differential Revision: D14954289
Pulled By: siying
fbshipit-source-id: 2c275255c1ac3992174e06529e44c55538325c94
Summary:
This is second attempt for #5101. Original commit message:
`BlockBasedTableIterator` avoid reading next block on `Next()` if it detects the iterator will be out of bound, by checking against index key. The optimization was added in #2239, and by the time it only check the bound per block. It seems later change make it a per-key check, which introduce unnecessary key comparisons.
This patch come with two fixes:
Fix 1: To optimize checking for bounds, we need comparing the bounds with index key as well. However BlockBasedTableIterator doesn't know whether its index iterator is internally using user keys or internal keys. The patch fixes that by extending InternalIterator with a user_key() function that is overridden by In IndexBlockIter.
Fix 2: In #5101 we return `IsOutOfBound()=true` when block index key is out of bound. But the index key can be larger than smallest key of the next file on the level. That file can be within upper bound and should not be filtered out.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5142
Differential Revision: D14907113
Pulled By: siying
fbshipit-source-id: ac95775c5b4e7b700f76ab43e39f45402c98fbfb
Summary:
Right now, two separate pieces of code are used to create WAL files in DBImpl::Open function of db_impl_open.cc and DBImpl::SwitchMemtable function of db_impl_write.cc. This code change simply creates 1 function called DBImpl::CreateWAL in db_impl_open.cc which is used to replace existing WAL creation logic in DBImpl::Open and DBImpl::SwitchMemtable.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5188
Differential Revision: D14942832
Pulled By: vjnadimpalli
fbshipit-source-id: d49230e04c36176015c8c1b422575872f92157fb
Summary:
Found this when test driving the new MultiGet. If you pass unsorted result with sorted_result = false you'll trigger the ASSERT incorrect even though we'll sort down below.
I've also added simple test cover sorted_result=true/false scenario copied from MultiGetSimple.
anand1976
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5195
Differential Revision: D14935475
Pulled By: yizhang82
fbshipit-source-id: 1d2af5e3a003847d965066a16e3b19da68acf170
Summary:
Before using prefix extractor `InDomain()` should be check. All uses in memtable.cc didn't check `InDomain()`.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5190
Differential Revision: D14923773
Pulled By: miasantreble
fbshipit-source-id: b3ad60bcca5f3a1a2b929a6eb34b0b7ba6326f04
Summary:
When ReadOption doesn't specify a snapshot, WritePrepared::Get used kMaxSequenceNumber to avoid the cost of creating a new snapshot object (that requires sync over db_mutex). This creates a race condition if it is reading from the writes of a transaction that had duplicate keys: each instance of duplicate key is inserted with a different sequence number and depending on the ordering the ::Get might skip the newer one and read the older one that is obsolete.
The patch fixes that by using last published seq as the snapshot sequence number. It also adds a check after the read is done to ensure that the max_evicted_seq has not advanced the aforementioned seq, which is a very unlikely event. If it did, then the read is not valid since the seq is not backed by an actually snapshot to let IsInSnapshot handle that properly when an overlapping commit is evicted from commit cache.
A unit test is added to reproduce the race condition with duplicate keys.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5147
Differential Revision: D14758815
Pulled By: maysamyabandeh
fbshipit-source-id: a56915657132cf6ba5e3f5ea1b5d78c803407719
Summary:
Since Statistics::measureTime() is deprecated, StatisticsImpl::measureTime() is not implemented. We realized that users might have a wrapped Statistics implementation in which measureTime() is implemented as forwarded to StatisticsImpl, and causes assert failure. In order to make the change less intrusive, we implement StatisticsImpl::measureTime(). We will revisit whether we need to remove it after several releases.
Also, add a test to make sure that a Statistics implementation using the old interface still works.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5181
Differential Revision: D14907089
Pulled By: siying
fbshipit-source-id: 29b6202fd04e30ed6f6adcaeb1000e87f10d1e1a
Summary:
When a new SST file is created via flush or compaction, we dump out the table properties, however only a few table properties are logged. The change here is to log all the table properties
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5168
Differential Revision: D14876928
Pulled By: vjnadimpalli
fbshipit-source-id: 1aca42ad00f9f650761d39e187f8beeb8700149b
Summary:
This PR introduces a new MultiGet() API, with the underlying implementation grouping keys based on SST file and batching lookups in a file. The reason for the new API is twofold - the definition allows callers to allocate storage for status and values on stack instead of std::vector, as well as return values as PinnableSlices in order to avoid copying, and it keeps the original MultiGet() implementation intact while we experiment with batching.
Batching is useful when there is some spatial locality to the keys being queries, as well as larger batch sizes. The main benefits are due to -
1. Fewer function calls, especially to BlockBasedTableReader::MultiGet() and FullFilterBlockReader::KeysMayMatch()
2. Bloom filter cachelines can be prefetched, hiding the cache miss latency
The next step is to optimize the binary searches in the level_storage_info, index blocks and data blocks, since we could reduce the number of key comparisons if the keys are relatively close to each other. The batching optimizations also need to be extended to other formats, such as PlainTable and filter formats. This also needs to be added to db_stress.
Benchmark results from db_bench for various batch size/locality of reference combinations are given below. Locality was simulated by offsetting the keys in a batch by a stride length. Each SST file is about 8.6MB uncompressed and key/value size is 16/100 uncompressed. To focus on the cpu benefit of batching, the runs were single threaded and bound to the same cpu to eliminate interference from other system events. The results show a 10-25% improvement in micros/op from smaller to larger batch sizes (4 - 32).
Batch Sizes
1 | 2 | 4 | 8 | 16 | 32
Random pattern (Stride length 0)
4.158 | 4.109 | 4.026 | 4.05 | 4.1 | 4.074 - Get
4.438 | 4.302 | 4.165 | 4.122 | 4.096 | 4.075 - MultiGet (no batching)
4.461 | 4.256 | 4.277 | 4.11 | 4.182 | 4.14 - MultiGet (w/ batching)
Good locality (Stride length 16)
4.048 | 3.659 | 3.248 | 2.99 | 2.84 | 2.753
4.429 | 3.728 | 3.406 | 3.053 | 2.911 | 2.781
4.452 | 3.45 | 2.833 | 2.451 | 2.233 | 2.135
Good locality (Stride length 256)
4.066 | 3.786 | 3.581 | 3.447 | 3.415 | 3.232
4.406 | 4.005 | 3.644 | 3.49 | 3.381 | 3.268
4.393 | 3.649 | 3.186 | 2.882 | 2.676 | 2.62
Medium locality (Stride length 4096)
4.012 | 3.922 | 3.768 | 3.61 | 3.582 | 3.555
4.364 | 4.057 | 3.791 | 3.65 | 3.57 | 3.465
4.479 | 3.758 | 3.316 | 3.077 | 2.959 | 2.891
dbbench command used (on a DB with 4 levels, 12 million keys)-
TEST_TMPDIR=/dev/shm numactl -C 10 ./db_bench.tmp -use_existing_db=true -benchmarks="readseq,multireadrandom" -write_buffer_size=4194304 -target_file_size_base=4194304 -max_bytes_for_level_base=16777216 -num=12000000 -reads=12000000 -duration=90 -threads=1 -compression_type=none -cache_size=4194304000 -batch_size=32 -disable_auto_compactions=true -bloom_bits=10 -cache_index_and_filter_blocks=true -pin_l0_filter_and_index_blocks_in_cache=true -multiread_batched=true -multiread_stride=4
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5011
Differential Revision: D14348703
Pulled By: anand1976
fbshipit-source-id: 774406dab3776d979c809522a67bedac6c17f84b
Summary:
Change the behavior of OptimizeForSmallDb() so that it is less likely to go out of memory.
Change the behavior of OptimizeForPointLookup() to take advantage of the new memtable whole key filter, and move away from prefix extractor as well as hash-based indexing, as they are prone to misuse.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5165
Differential Revision: D14880709
Pulled By: siying
fbshipit-source-id: 9af30e3c9e151eceea6d6b38701a58f1f9fb692d
Summary:
Introducing Periodic Compactions.
This feature allows all the files in a CF to be periodically compacted. It could help in catching any corruptions that could creep into the DB proactively as every file is constantly getting re-compacted. And also, of course, it helps to cleanup data older than certain threshold.
- Introduced a new option `periodic_compaction_time` to control how long a file can live without being compacted in a CF.
- This works across all levels.
- The files are put in the same level after going through the compaction. (Related files in the same level are picked up as `ExpandInputstoCleanCut` is used).
- Compaction filters, if any, are invoked as usual.
- A new table property, `file_creation_time`, is introduced to implement this feature. This property is set to the time at which the SST file was created (and that time is given by the underlying Env/OS).
This feature can be enabled on its own, or in conjunction with `ttl`. It is possible to set a different time threshold for the bottom level when used in conjunction with ttl. Since `ttl` works only on 0 to last but one levels, you could set `ttl` to, say, 1 day, and `periodic_compaction_time` to, say, 7 days. Since `ttl < periodic_compaction_time` all files in last but one levels keep getting picked up based on ttl, and almost never based on periodic_compaction_time. The files in the bottom level get picked up for compaction based on `periodic_compaction_time`.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5166
Differential Revision: D14884441
Pulled By: sagar0
fbshipit-source-id: 408426cbacb409c06386a98632dcf90bfa1bda47
Summary:
Create new function NPHash64() and GetSliceNPHash64(), which are currently
implemented using murmurhash.
Replace the current direct call of murmurhash() to use the new functions
if the hash results are not used in on-disk format.
This will make it easier to try out or switch to alternative functions
in the uses where data format compatibility doesn't need to be considered.
This part shouldn't have any performance impact.
Also, the sharded cache hash function is changed to the new format, because
it falls into this categoery. It doesn't show visible performance impact
in db_bench results. CPU showed by perf is increased from about 0.2% to 0.4%
in an extreme benchmark setting (4KB blocks, no-compression, everything
cached in block cache). We've known that the current hash function used,
our own Hash() has serious hash quality problem. It can generate a lots of
conflicts with similar input. In this use case, it means extra lock contention
for reads from the same file. This slight CPU regression is worthy to me
to counter the potential bad performance with hot keys. And hopefully this
will get further improved in the future with a better hash function.
cache_test's condition is relaxed a little bit to. The new hash is slightly
more skewed in this use case, but I manually checked the data and see
the hash results are still in a reasonable range.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5155
Differential Revision: D14834821
Pulled By: siying
fbshipit-source-id: ec9a2c0a2f8ae4b54d08b13a5c2e9cc97aa80cb5
Summary:
Expose DB methods to lock and unlock the WAL.
These methods are intended to use by MyRocks in order to obtain WAL
coordinates in consistent way.
Usage scenario is following:
MySQL has performance_schema.log_status which provides information that
enables a backup tool to copy the required log files without locking for
the duration of copy. To populate this table MySQL does following:
1. Lock the binary log. Transactions are not allowed to commit now
2. Save the binary log coordinates
3. Walk through the storage engines and lock writes on each engine. For
InnoDB, redo log is locked. For MyRocks, WAL should be locked.
4. Ask storage engines for their coordinates. InnoDB reports its current
LSN and checkpoint LSN. MyRocks should report active WAL files names
and sizes.
5. Release storage engine's locks
6. Unlock binary log
Backup tool will then use this information to copy InnoDB, RocksDB and
MySQL binary logs up to specified positions to end up with consistent DB
state after restore.
Currently, RocksDB allows to obtain the list of WAL files. Only missing
bit is the method to lock the writes to WAL files.
LockWAL method must flush the WAL in order for the reported size to be
accurate (GetSortedWALFiles is using file system stat call to return the
file size), also, since backup tool is going to copy the WAL, it is
better to be flushed.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5146
Differential Revision: D14815447
Pulled By: maysamyabandeh
fbshipit-source-id: eec9535a6025229ed471119f19fe7b3d8ae888a3
Summary:
Annotate all of the logging functions to inform the compiler that these
use printf-style formatting arguments. This allows the compiler to emit
warnings if the format arguments are incorrect.
This also fixes many problems reported now that format string checking
is enabled. Many of these are simply mix-ups in the argument type (e.g,
int vs uint64_t), but in several cases the wrong number of arguments
were being passed in which can cause the code to crash.
The primary motivation for this was to fix the log message in
`DBImpl::SwitchMemtable()` which caused a segfault due to an extra %s
format parameter with no argument supplied.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5089
Differential Revision: D14574795
Pulled By: simpkins
fbshipit-source-id: 0921b03f0743652bf4ae21e414ff54b3bb65422a
Summary:
The ReadCallback was shared between all threads in IteratorWithLocalStatistics. A race condition was
hence introduced with recent changes that changes the content of ReadCallback. The patch fixes that by using a separate callback per thread.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5149
Differential Revision: D14761612
Pulled By: maysamyabandeh
fbshipit-source-id: 814a316aed046c318cb90e22379a6e32ac528949
Summary:
Although user should first call StartTrace to begin the RocksDB tracing function and call EndTrace to stop the tracing process, user can accidentally call EndTrace first. It will cause segment fault and crash the DB instance. The issue is fixed by checking the pointer first.
Test case added in db_test2.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5130
Differential Revision: D14691420
Pulled By: zhichao-cao
fbshipit-source-id: 3be13d2f944bc453728ef8eef67b68d7ad0939c8
Summary:
This PR address two open issues:
1. clang analyzer is paranoid about db_ being nullptr after DB::Open calls in the test.
See https://github.com/facebook/rocksdb/pull/5043#discussion_r271394579
Add an assert to keep clang happy
2. PR https://github.com/facebook/rocksdb/pull/5049 introduced a variable shadowing:
```
db/db_iterator_test.cc: In constructor ‘rocksdb::DBIteratorWithReadCallbackTest_ReadCallback_Test::TestBody()::TestReadCallback::TestReadCallback(rocksdb::SequenceNumber)’:
db/db_iterator_test.cc:2484:9: error: declaration of ‘max_visible_seq’ shadows a member of 'this' [-Werror=shadow]
: ReadCallback(max_visible_seq) {}
^
```
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5140
Differential Revision: D14735497
Pulled By: miasantreble
fbshipit-source-id: 3219ea75cf4ae04f64d889323f6779e84be98144
Summary:
In prepare phase of 2PC, the db promises to remember the prepared data, for possible future commits. To fulfill the promise the prepared data must be persisted in the WAL so that they could be recovered after a crash. The log that contains a prepare batch that is not committed yet, is marked so that it is not garbage collected before the transaction commits/rollbacks. The bug was that the write to the log file and the mark of the file was not atomic, and WAL gc could have happened before the WAL log is actually marked. This patch moves the marking logic to PreReleaseCallback so that the WAL gc logic that joins both write threads would see the WAL write and WAL mark atomically.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5121
Differential Revision: D14665210
Pulled By: maysamyabandeh
fbshipit-source-id: 1d66aeb1c66a296cb4899a5a20c4d40c59e4b534
Summary:
WriteUnPrepared adds a virtual function, MaxUnpreparedSequenceNumber, to ReadCallback, which returns 0 unless WriteUnPrepared is enabled and the transaction has uncommitted data written to the DB. Together with snapshot sequence number, this determines the last sequence that is visible to reads.
The patch clarifies the guarantees of the GetIterator API in WriteUnPrepared transactions and make use of that to statically initialize the read callback and thus avoid the virtual call.
Furthermore it increases the minimum value for min_uncommitted from 0 to 1 as seq 0 is used only for last level keys that are committed in all snapshots.
The following benchmark shows +0.26% higher throughput in seekrandom benchmark.
Benchmark:
./db_bench --benchmarks=fillrandom --use_existing_db=0 --num=1000000 --db=/dev/shm/dbbench
./db_bench --benchmarks=seekrandom[X10] --use_existing_db=1 --db=/dev/shm/dbbench --num=1000000 --duration=60 --seek_nexts=100
seekrandom [AVG 10 runs] : 20355 ops/sec; 225.2 MB/sec
seekrandom [MEDIAN 10 runs] : 20425 ops/sec; 225.9 MB/sec
./db_bench_lessvirtual3 --benchmarks=seekrandom[X10] --use_existing_db=1 --db=/dev/shm/dbbench --num=1000000 --duration=60 --seek_nexts=100
seekrandom [AVG 10 runs] : 20409 ops/sec; 225.8 MB/sec
seekrandom [MEDIAN 10 runs] : 20487 ops/sec; 226.6 MB/sec
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5049
Differential Revision: D14366459
Pulled By: maysamyabandeh
fbshipit-source-id: ebaff8908332a5ae9af7defeadabcb624be660ef
Summary:
This reverts commit f29dc1b906.
In BlockBasedTableIterator, index_iter_->key() is sometimes a user key, so it is wrong to call ExtractUserKey() against it. This is a bug introduced by #5101.
Temporarily revert the diff to keep the branch clean.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5132
Differential Revision: D14718584
Pulled By: siying
fbshipit-source-id: 0ac55dc9b5dbc18c7809092146bdf7eb9364b9ad
Summary:
Just like ReadOptions::background_purge_on_iterator_cleanup but for ColumnFamilyHandle instead of Iterator.
In our use case we sometimes call ColumnFamilyHandle's destructor from low-latency threads, and sometimes it blocks the thread for a few seconds deleting the files. To avoid that, we can either offload ColumnFamilyHandle's destruction to a background thread on our side, or add this option on rocksdb side. This PR does the latter, to be consistent with how we solve exactly the same problem for iterators using background_purge_on_iterator_cleanup option.
(EDIT: It's avoid_unnecessary_blocking_io now, and affects both CF drops and iterator destructors.)
I'm not quite comfortable with having two separate options (background_purge_on_iterator_cleanup and background_purge_on_cf_cleanup) for such a rarely used thing. Maybe we should merge them? Rename background_purge_on_cf_cleanup to something like delete_files_on_background_threads_only or avoid_blocking_io_in_unexpected_places, and make iterators use it instead of the one in ReadOptions? I can do that here if you guys think it's better.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5043
Differential Revision: D14339233
Pulled By: al13n321
fbshipit-source-id: ccf7efa11c85c9a5b91d969bb55627d0fb01e7b8
Summary:
`BlockBasedTableIterator` avoid reading next block on `Next()` if it detects the iterator will be out of bound, by checking against index key. The optimization was added in #2239, and by the time it only check the bound per block. It seems later change make it a per-key check, which introduce unnecessary key comparisons.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5101
Differential Revision: D14678707
Pulled By: siying
fbshipit-source-id: 2372446116753c7892ea4cec7b4b49ef87ba463e
Summary:
**This PR updates RepeatableThread::wait, breaking some tests on OS X. The rest of the PR fixes the tests on OS X.**
`RepeatableThreadTest.MockEnvTest` uses `MockTimeEnv` and `RepeatableThread`. If `RepeatableThread::wait` calls `TimedWait` with a time smaller than or equal to the current (real) time, `TimedWait` returns immediately on certain platforms, e.g. OS X. #4560 addresses this issue by replacing `TimedWait` with `Wait` in test. This fixes the test but makes test/production code diverge, which is not optimal for test coverage. This PR proposes an alternative fix which unifies test and production code path for `RepeatableThread::wait`. We obtain the current (real) time in seconds and add 10 extra seconds to ensure that `RepeatableThread::wait` invokes `TimedWait` with a time greater than (real) current time. This is to prevent the `TimedWait` function from returning immediately without sleeping and releasing the mutex. If `TimedWait` returns immediately, the mutex will not be released, and `RepeatableThread::TEST_WaitForRun` never has a chance to execute the callback which, in this case, updates the result returned by `mock_env->NowMicros()`. Consequently, `RepeatableThread::wait` cannot break out of the loop, causing test to hang. The extra 10 seconds is a best-effort approach because there seems no reliable and deterministic way to provide the aforementioned guarantee. By the time `RepeatableThread::wait` is called, there is no guarantee that the `delay + mock_env->NowMicros()` will be greater than the current real time. However, 10 seconds should be sufficient in most cases. We will keep an eye for possible flakiness of this test.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5107
Differential Revision: D14680885
Pulled By: riversand963
fbshipit-source-id: d1ecbe10e1dacd110bd464cd01e188bfee72b89e
Summary:
WAL files are currently not subject to deletion rate limiting by DeleteScheduler. If the size of the WAL files is significant, this can cause a high delete rate on SSDs that may affect other operations. To fix it, force WAL file deletions to go through the SstFileManager. Original PR for this is #2768
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5116
Differential Revision: D14669437
Pulled By: anand1976
fbshipit-source-id: c5f62d0640cebaa1574de841a1d01e4ce2faadf0
Summary:
We see a failure of obsolete_files_test but aren't able to identify
the issue. Improve the test in following way and hope we can debug
better next time:
1. Place sync point before automatic compaction runs so race condition
will always trigger.
2. Disable sync point before test finishes.
3. ASSERT_OK() instead of ASSERT_TRUE(status.ok())
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5125
Differential Revision: D14669456
Pulled By: siying
fbshipit-source-id: dccb7648e334501ad651eb212880096eef1f4ab2
Summary:
Following files were run through automatic formatter:
db/db_impl.cc
db/db_impl.h
db/db_impl_compaction_flush.cc
db/db_impl_debug.cc
db/db_impl_files.cc
db/db_impl_readonly.h
db/db_impl_write.cc
db/dbformat.cc
db/dbformat.h
table/block.cc
table/block.h
table/block_based_filter_block.cc
table/block_based_filter_block.h
table/block_based_filter_block_test.cc
table/block_based_table_builder.cc
table/block_based_table_reader.cc
table/block_based_table_reader.h
table/block_builder.cc
table/block_builder.h
table/block_fetcher.cc
table/block_prefix_index.cc
table/block_prefix_index.h
table/block_test.cc
table/format.cc
table/format.h
I could easily run all the files, but I don't want people to feel that
I'm doing it for lines of code changes :)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5114
Differential Revision: D14633040
Pulled By: siying
fbshipit-source-id: 3f346cb53bf21e8c10704400da548dfce1e89a52
Summary:
Currently `perf_context.user_key_comparison_count` is bump only in `InternalKeyComparator`. For places user comparator is used directly the counter is not bump. Fixing the majority of it.
Index iterator and filter code also use user comparator directly and don't bump the counter. It is not fixed in this patch.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5098
Differential Revision: D14603753
Pulled By: siying
fbshipit-source-id: 1cd41035644ca9e49b97a51030a5d1e15f5f3cae
Summary:
The code convention we are following, Google C++ Style, discourage
alias in header files, especially public headers:
https://google.github.io/styleguide/cppguide.html#Aliases
Remove some of them. Might removed some from .cc files as well to be consistent.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5113
Differential Revision: D14633030
Pulled By: siying
fbshipit-source-id: b990edc919d5de60295992284f980195e501d424
Summary:
This PR allows RocksDB to run in single-primary, multi-secondary process mode.
The writer is a regular RocksDB (e.g. an `DBImpl`) instance playing the role of a primary.
Multiple `DBImplSecondary` processes (secondaries) share the same set of SST files, MANIFEST, WAL files with the primary. Secondaries tail the MANIFEST of the primary and apply updates to their own in-memory state of the file system, e.g. `VersionStorageInfo`.
This PR has several components:
1. (Originally in #4745). Add a `PathNotFound` subcode to `IOError` to denote the failure when a secondary tries to open a file which has been deleted by the primary.
2. (Similar to #4602). Add `FragmentBufferedReader` to handle partially-read, trailing record at the end of a log from where future read can continue.
3. (Originally in #4710 and #4820). Add implementation of the secondary, i.e. `DBImplSecondary`.
3.1 Tail the primary's MANIFEST during recovery.
3.2 Tail the primary's MANIFEST during normal processing by calling `ReadAndApply`.
3.3 Tailing WAL will be in a future PR.
4. Add an example in 'examples/multi_processes_example.cc' to demonstrate the usage of secondary RocksDB instance in a multi-process setting. Instructions to run the example can be found at the beginning of the source code.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4899
Differential Revision: D14510945
Pulled By: riversand963
fbshipit-source-id: 4ac1c5693e6012ad23f7b4b42d3c374fecbe8886
Summary:
Introduce CPU timers for iterator seek and next operations. Seek
counter includes SeekToFirst, SeekToLast and SeekForPrev, w/ the
caveat that SeekToLast timer doesn't include some post processing
time if upper bound is defined.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5076
Differential Revision: D14525218
Pulled By: fredfsh
fbshipit-source-id: 03ba25df3b22b06c072621e4de0eacfa1445f0d9
Summary:
With https://github.com/facebook/rocksdb/pull/3009 we go through every CF
to check whether a bottommost compaction is needed to be triggered. This is done
within DB mutex. What we do within DB mutex may heavily influece the write throughput
we can achieve, so we always want to minimize work there.
Here we try to avoid this for-loop by first check a global threshold. In most of
the time, the CF loop can be avoided.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5090
Differential Revision: D14582684
Pulled By: siying
fbshipit-source-id: 968f6d9bb6affe1a5ebc4910b418300b076f166f
Summary:
The patch reorders DBIter fields to put 1-byte fields together and let the compiler optimize the memory usage by using less 64-bit allocations for bools and enums.
This might have a negative side effect of putting the variables that are accessed together into different cache lines and hence increasing the cache misses. Not sure what benchmark would verify that thought. I ran simple, single-threaded seekrandom benchmarks but the variance in the results is too much to be conclusive.
./db_bench --benchmarks=fillrandom --use_existing_db=0 --num=1000000 --db=/dev/shm/dbbench
./db_bench --benchmarks=seekrandom[X10] --use_existing_db=1 --db=/dev/shm/dbbench --num=1000000 --duration=60 --seek_nexts=100
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5078
Differential Revision: D14562676
Pulled By: maysamyabandeh
fbshipit-source-id: 2284655d46e079b6e9a860e94be5defb6f482167
Summary:
Add an option to filter out READ or WRITE operations while tracing.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5082
Differential Revision: D14515083
Pulled By: mrmiywj
fbshipit-source-id: 2504c89a9abf1dd629cad44b4104092702d77610
Summary:
This is a feature to sample data-block compressibility and and report them as stats. 1 in N (tunable) blocks is sampled for compressibility using two algorithms:
1. lz4 or snappy for fast compression
2. zstd or zlib for slow but higher compression.
The stats are reported to the caller as raw-bytes and compressed-bytes. The block continues to be compressed for storage using the specified CompressionType.
The db_bench_tool how has a command line option for specifying the sampling rate. It's default value is 0 (no sampling). To test the overhead for a certain value, users can compare the performance of db_bench_tool, varying the sampling rate. It is unlikely to have a noticeable impact for high values like 20.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4842
Differential Revision: D13629011
Pulled By: shobhitdayal
fbshipit-source-id: 14ca668bcab6499b2a1734edf848eb62a4f4fafa
Summary:
There is a potential failure case in DBImpl::SwitchMemtable() that is not handled properly. The call to cur_log_writer->WriteBuffer() can fail due to an IO error. In that case, we need to call SetBGError() in order set the background error since the WriteBuffer() failure may result in data loss.
Also, the asserts for !new_mem and !new_log are incorrect, as those would have been allocated by the time this failure is detected.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5072
Differential Revision: D14461384
Pulled By: anand1976
fbshipit-source-id: fb59bce9d61378f37d2dfcd28c0b704b0f43c3cf
Summary:
This reverts commit ee1818081f.
We are not ready to deprecate this feature. revert it for now.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5034
Differential Revision: D14287246
Pulled By: siying
fbshipit-source-id: e4beafdeaee1c94364fdaa6ba198218d158339f7
Summary:
Right now, users can change statistics.stats_level while DB is running, but TSAN may report
data race. We make stats_level_ to be atomic, and access them using accessors.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5030
Differential Revision: D14267519
Pulled By: siying
fbshipit-source-id: 37d7ebeff7a43a406230143422a16af899163f73
Summary:
PreReleaseCallback meant to be called before the writes are visible to the readers. Since the sequence number is known after the WAL write, there is no reason to delay calling PreReleaseCallback to after the memtable write, which would complicates the reader's logic in presence of our memtable writes that are made visible by the other write thread.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5015
Differential Revision: D14221670
Pulled By: maysamyabandeh
fbshipit-source-id: a504dd665cf923226d7af09cc8e9c7739a25edc6
Summary:
Statistics cost too much CPU for some use cases. Add two stats levels
so that people can choose to skip two types of expensive stats, timers and
histograms.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5027
Differential Revision: D14252765
Pulled By: siying
fbshipit-source-id: 75ecec9eaa44c06118229df4f80c366115346592
Summary:
The read path includes a callback function, ReadCallback, which would eventually calls IsInSnapshot to figure if a particular seq is in the reading snapshot or not. This callback is virtual, which adds the cost of multiple virtual function call to each read. The first few checks in IsInSnapshot, however, are quite trivial and take care of majority of the cases. The patch moves those to a non-virtual function in the the parent class, ReadCallback, to lower the virtual callback cost.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5018
Differential Revision: D14226562
Pulled By: maysamyabandeh
fbshipit-source-id: 6feed5b34f3b082e52092c5ef143e29b49c46b44
Summary:
This PR adds public `GetStatsHistory` API to retrieve stats history in the form of an std map. The key of the map is the timestamp in microseconds when the stats snapshot is taken, the value is another std map from stats name to stats value (stored in std string). Two DBOptions are introduced: `stats_persist_period_sec` (default 10 minutes) controls the intervals between two snapshots are taken; `max_stats_history_count` (default 10) controls the max number of history snapshots to keep in memory. RocksDB will stop collecting stats snapshots if `stats_persist_period_sec` is set to 0.
(This PR is the in-memory part of https://github.com/facebook/rocksdb/pull/4535)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4748
Differential Revision: D13961471
Pulled By: miasantreble
fbshipit-source-id: ac836d401ecb84ea92216bf9966f969dedf4ad04
Summary:
MyRocks calls `GetForUpdate` on `INSERT`, for unique key check, and in almost all cases GetForUpdate returns empty result. For such cases, whole key bloom filter is helpful.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4985
Differential Revision: D14118257
Pulled By: miasantreble
fbshipit-source-id: d35cb7109c62fd5ad541a26968e3a3e16d3e85ea
Summary:
The info log header feature never worked well, because log level Header was not
translated to Logger::LogHeader() call. Fix it.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4980
Differential Revision: D14087283
Pulled By: siying
fbshipit-source-id: 7e7d03ce35fa8d13d4ee549f46f7326f7bc0006d
Summary:
Right now when a flush is triggered, the memory consumption is logged but data size is not.
It's useful to log both when we debug unexpected small flushed file size.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4979
Differential Revision: D14071979
Pulled By: siying
fbshipit-source-id: 0cd60449c5205eb00e0fbc299084418f609904ed
Summary:
We introduced ttl option in CompactionOptionsFIFO when ttl-based file
deletion (compaction) was supported only as part of FIFO Compaction. But
with the extension of ttl semantics even to Level compaction,
CompactionOptionsFIFO.ttl can now be deprecated. Instead we will start
using ColumnFamilyOptions.ttl for FIFO compaction as well.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4965
Differential Revision: D14072960
Pulled By: sagar0
fbshipit-source-id: c98cc2ae695a28136295787cd88d36a220fc219e
Summary:
If `CompressionOptions::max_dict_bytes` and/or `CompressionOptions::zstd_max_train_bytes` are set, `SstFileWriter` will now generate files respecting those options.
I refactored the logic a bit for deciding when to use dictionary compression. Previously we plumbed `is_bottommost_level` down to the table builder and used that. However it was kind of confusing in `SstFileWriter`'s context since we don't know what level the file will be ingested to. Instead, now the higher-level callers (e.g., flush, compaction, file writer) are responsible for building the right `CompressionOptions` to give the table builder.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4978
Differential Revision: D14060763
Pulled By: ajkr
fbshipit-source-id: dc802c327896df2b319dc162d6acc82b9cdb452a
Summary:
if an operation just involves a single column family, then we do
not have to set the kInAtomicGroup tag when writing to MANIFEST. This change
can fix a compatibility test failure, i.e. 5.15 and earlier cannot recognize
kInAtomicGroup tag.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4981
Differential Revision: D14072687
Pulled By: riversand963
fbshipit-source-id: 46b0c61e399f16c6b7169de0b33430d0ed90d6d4
Summary:
Make file ingestion atomic.
as title.
Ingesting external SST files into multiple column families should be atomic. If
a crash occurs and db reopens, either all column families have successfully
ingested the files before the crash, or non of the ingestions have any effect
on the state of the db.
Also add unit tests for atomic ingestion.
Note that the unit test here does not cover the case of incomplete atomic group
in the MANIFEST, which is covered in VersionSetTest already.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4895
Differential Revision: D13718245
Pulled By: riversand963
fbshipit-source-id: 7df97cc483af73ad44dd6993008f99b083852198
Summary:
Previously, stats were logged in warning level. This was done in that way because
people reported that it wasn't logged in MyRocks. However, later we learned that it turns
out to be due to a bug in MyRocks, which is fixed in
79bb705e74
Now we revert the stats logging to INFO level, so that it doesn't pollute the warning
level logging.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4977
Differential Revision: D14058485
Pulled By: siying
fbshipit-source-id: 19fab323c19d9bc88184287f209551f9a77ca0e6
Summary:
In `DBImpl::AtomicFlushMemTablesToOutputFiles`, we need to call fsync only once
on the same data directory. If two column families share a common directory for
their data, we call fsync only once.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4817
Differential Revision: D13543689
Pulled By: riversand963
fbshipit-source-id: 4701d77c96a47802fbf6cb9f3337ee65d46b95f5
Summary:
Our previous approach was to train one compression dictionary per compaction, using the first output SST to train a dictionary, and then applying it on subsequent SSTs in the same compaction. While this was great for minimizing CPU/memory/I/O overhead, it did not achieve good compression ratios in practice. In our most promising potential use case, moderate reductions in a dictionary's scope make a major difference on compression ratio.
So, this PR changes compression dictionary to be scoped per-SST. It accepts the tradeoff during table building to use more memory and CPU. Important changes include:
- The `BlockBasedTableBuilder` has a new state when dictionary compression is in-use: `kBuffered`. In that state it accumulates uncompressed data in-memory whenever `Add` is called.
- After accumulating target file size bytes or calling `BlockBasedTableBuilder::Finish`, a `BlockBasedTableBuilder` moves to the `kUnbuffered` state. The transition (`EnterUnbuffered()`) involves sampling the buffered data, training a dictionary, and compressing/writing out all buffered data. In the `kUnbuffered` state, a `BlockBasedTableBuilder` behaves the same as before -- blocks are compressed/written out as soon as they fill up.
- Samples are now whole uncompressed data blocks, except the final sample may be a partial data block so we don't breach the user's configured `max_dict_bytes` or `zstd_max_train_bytes`. The dictionary trainer is supposed to work better when we pass it real units of compression. Previously we were passing 64-byte KV samples which was not realistic.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4952
Differential Revision: D13967980
Pulled By: ajkr
fbshipit-source-id: 82bea6f7537e1529c7a1a4cdee84585f5949300f
Summary:
If IsInSnapshot(seq2, snapshot) determines that the snapshot is released, the future queries IsInSnapshot(seq1, snapshot) could still return a definitive answer of true if for example seq1 is too old that is determined visible in all snapshots. This violates a recently added assert statement to compaction iterator. The patch relaxes the assert.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4969
Differential Revision: D14030998
Pulled By: maysamyabandeh
fbshipit-source-id: 6db53db0e37d0a20e8997ef2c1004b8627614ab9
Summary:
Always enable properties block checksum verification for block-based table. For external SST file ingested with 'write_global_seqno==true', we use 'DecodeEntrySlow' to parse its blocks' contents so that the process will not die upon failing the assertion possibly caused by corruption.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4956
Differential Revision: D14012741
Pulled By: riversand963
fbshipit-source-id: 8b766e6f54b36f8f9e074c0e19e0926ec3cce186
Summary:
Implement trace sampling to allow user to specify the sampling frequency, i.e. save one per how many requests, so that a user does not need to log all if he/she is interested in only a sampled set.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4963
Differential Revision: D14011190
Pulled By: tang-jianfeng
fbshipit-source-id: 078b631d9319b67cb089dd2c30e21d0df8dc406a
Summary:
We want to reserve some right that some extra information added manifest
in the future can be forward compatible by previous versions. Now we create a
place holder for that. A bit in tag is added to indicate that a field can be
safely ignored.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4960
Differential Revision: D14000484
Pulled By: siying
fbshipit-source-id: cbf5bad3f9d5ec798f789806f244d1c20d3b66d6
Summary:
We found that the behavior of CompactionFilter::IgnoreSnapshots() = false isn't
what we have expected. We thought that snapshot will always be preserved.
However, we just realized that, if no snapshot is created while compaction
starts, and a snapshot is created after that, the data seen from the snapshot
can successfully be dropped by the compaction. This creates a strange behavior
to the feature, which is hard to explain. Like what is documented in code
comment, this feature is not very useful with snapshot anyway. The decision
is to deprecate the feature.
We keep the function to avoid to break users code. However, we will fail
compactions if false is returned.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4954
Differential Revision: D13981900
Pulled By: siying
fbshipit-source-id: 2db8c2c3865acd86a28dca625945d1481b1d1e36
Summary:
Cuckoo Hash is less useful than we initially expected. Remove it.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4953
Differential Revision: D13979264
Pulled By: siying
fbshipit-source-id: 2a60afdaa989f045357398b43a1cc5d46f4492ed
Summary:
This will fix the following build error:
> db/db_test.cc: In member function ‘virtual void rocksdb::DBTest_CompactFilesShouldTriggerAutoCompaction_Test::TestBody()’:
> db/db_test.cc:5462:8: error: ‘class rocksdb::DB’ has no member named ‘GetColumnFamilyMetaData’
> db_->GetColumnFamilyMetaData(db_->DefaultColumnFamily(), &cf_meta_data);
> db/db_test.cc:5490:8: error: ‘class rocksdb::DB’ has no member named ‘GetColumnFamilyMetaData’
> db_->GetColumnFamilyMetaData(db_->DefaultColumnFamily(), &cf_meta_data);
> db/db_test.cc:5499:8: error: ‘class rocksdb::DB’ has no member named ‘GetColumnFamilyMetaData’
> db_->GetColumnFamilyMetaData(db_->DefaultColumnFamily(), &cf_meta_data);
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4950
Differential Revision: D13965378
Pulled By: miasantreble
fbshipit-source-id: a975435476fe555b1cd9d5da263ee3da3acdea56
Summary:
Existing implementation of PerfContext does not define copy constructor or assignment operator, which could potentially cause problems when user create copies and resets the builtin one. This PR address the issue by providing these two constructors with deep copy semantics.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4919
Differential Revision: D13960406
Pulled By: miasantreble
fbshipit-source-id: 36aab5aaee65d4480f537e4e22148faa45e8e334
Summary:
CompactFiles() may block auto compaction which could cuase DB hang when it
reachs level0_stop_writes_trigger.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4940
Differential Revision: D13929648
Pulled By: cooldoger
fbshipit-source-id: 10842df38df3bebf862cd1a120a88ce961fdd381
Summary:
In NotFound cases, stats BYTES_READ and perf_context.get_read_bytes is still be increased. The amount increased will be
whatever size of the string or PinnableSlice that users passed in as the output data structure. This is wrong. Fix this by not
increasing these two counters.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4938
Differential Revision: D13908963
Pulled By: siying
fbshipit-source-id: 60bce42e4fbb9862bba3da36dbc27b2963ea6162
Summary:
Fix the ouput overlap bug when using subcompactions, the upper bound of output
file was extended incorrectly.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4898
Differential Revision: D13736107
Pulled By: ajkr
fbshipit-source-id: 21dca09f81d5f07bf2766bf566f9b50dcab7d8e3
Summary:
The patch fixes the following analyze error by checking the return status of ParseInternalKey.
```
db/merge_helper.cc:306:23: warning: The right operand of '==' is a garbage value
assert(kTypeMerge == orig_ikey.type);
```
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4937
Differential Revision: D13908506
Pulled By: maysamyabandeh
fbshipit-source-id: 68d7771e75519da3d4bd807fd231675ec12093f6
Summary:
1. this commit fixes our handling of a combination of two separate edge
cases. If a flush job does not pick any memtable to flush (because another
flush job has already picked the same memtables), and the column family
assigned to the flush job is dropped right before RocksDB calls
rocksdb::InstallMemtableAtomicFlushResults, our original code passes
a FileMetaData object whose file number is 0, failing the assertion in
rocksdb::InstallMemtableAtomicFlushResults (assert(m->GetFileNumber() > 0)).
2. Also piggyback a small change: since we already create a local copy of column family's mutable CF options to eliminate potential race condition with `SetOptions` call, we might as well use the local copy in other function calls in the same scope.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4932
Differential Revision: D13901322
Pulled By: riversand963
fbshipit-source-id: b936580af7c127ea0c6c19ea10cd5fcede9fb0f9
Summary:
FlushMemTablesToOutputFiles calls FlushMemTableToOutputFile for each column family. The patch moves the take-snapshot logic to outside FlushMemTableToOutputFile so that it does it once for all the flushes. This also addresses a deadlock issue for resetting the managed snapshot of job_snapshot in the 2nd call to FlushMemTableToOutputFile.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4934
Differential Revision: D13900747
Pulled By: maysamyabandeh
fbshipit-source-id: f3cd650c5fff24cf95c1aaf8a10c149d42bf042c
Summary:
Measure CPU time consumed for a compaction and report it in the stats report
Enable NowCPUNanos() to work for MacOS
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4889
Differential Revision: D13701276
Pulled By: zinoale
fbshipit-source-id: 5024e5bbccd4dd10fd90d947870237f436445055
Summary:
before file ingestion (in preparation phase), verify the checksums of
the blocks of the external SST file, including properties block with global
seqno.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4916
Differential Revision: D13863501
Pulled By: riversand963
fbshipit-source-id: dc54697f970e3807832e2460f7228fcc7efe81ee
Summary:
While stepping through the code I noticed that there is a redundant call to TableFileName.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4925
Differential Revision: D13845749
Pulled By: sagar0
fbshipit-source-id: 31db45716b4d720e0e0350dd457b49d6f1848e7d
Summary:
Store_index_in_file is a less useful feature. To simplify the code to maintain, we are dropping the feature.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4914
Differential Revision: D13791883
Pulled By: siying
fbshipit-source-id: d187c5d662584866103e4b77d09dfb925509ae2e
Summary:
https://github.com/facebook/rocksdb/pull/4053 avoids memcopy for Get() results if files are immortable
(read-only DB, max_open_files=-1) and the file is ammaped. The same optimization is being applied to PlainTable
here.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4924
Differential Revision: D13827749
Pulled By: siying
fbshipit-source-id: 1f2cbfc530b40ce08ccd53f95f6e78de4d1c2f96
Summary:
I didn't find where customized hash function is used in DynamicBloom. This can only reduce performance. Remove it.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4915
Differential Revision: D13794452
Pulled By: siying
fbshipit-source-id: e38669b11e01444d2d782da11c7decabbd851819
Summary:
Previously compaction was not collapsing operands for a first
key on a layer, even in cases when it was its root of history. Some
tests (CompactionJobTest.NonAssocMerge) was actually accounting
for that bug,
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4909
Differential Revision: D13781169
Pulled By: finik
fbshipit-source-id: d2de353ecf05bec39b942cd8d5b97a8dc445f336
Summary:
- If block cache disabled or not used for meta-blocks, `BlockBasedTableReader::Rep::uncompression_dict` owns the `UncompressionDict`. It is preloaded during `PrefetchIndexAndFilterBlocks`.
- If block cache is enabled and used for meta-blocks, block cache owns the `UncompressionDict`, which holds dictionary and digested dictionary when needed. It is never prefetched though there is a TODO for this in the code. The cache key is simply the compression dictionary block handle.
- New stats for compression dictionary accesses in block cache: "BLOCK_CACHE_COMPRESSION_DICT_*" and "compression_dict_block_read_count"
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4881
Differential Revision: D13663801
Pulled By: ajkr
fbshipit-source-id: bdcc54044e180855cdcc57639b493b0e016c9a3f
Summary:
Even one key falls in a file's range, we can not infer it definitely exists in this file.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4902
Differential Revision: D13795018
Pulled By: siying
fbshipit-source-id: 590956f727e9440fcdee55ad9541ace934c64914
Summary:
compaction_pri = kMinOverlappingRatio usually provides much better write amplification than the default.
https://github.com/facebook/rocksdb/pull/4907 fixes one shortcome of this option. Make it default.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4911
Differential Revision: D13789262
Pulled By: siying
fbshipit-source-id: d90acf8c4dede44f00d183ca4c7a210259378269
Summary:
Right now, CompactionPri = kMinOverlappingRatio provides best write amplification, but it doesn't
prioritize files with more tombstones. We combine the two good features: make kMinOverlappingRatio
to boost files with lots of tombstones too.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4907
Differential Revision: D13788774
Pulled By: siying
fbshipit-source-id: 1991cbb495fb76c8b529de69896e38d81ed9d9b3
Summary:
This is essentially a re-submission of #4251 with a few improvements:
- Split `CompressionDict` into two separate classes: `CompressionDict` and `UncompressionDict`
- Eliminated `Init` functions. Instead do all initialization work in constructors.
- Added test case for parallel DB open, which is the scenario where #4251 failed under TSAN.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4849
Differential Revision: D13606039
Pulled By: ajkr
fbshipit-source-id: 08c236059798c710db9cbf545fce0f371232d447
Summary:
Fix how CompactionIterator::findEarliestVisibleSnapshots handles released snapshot. It fixing the two scenarios:
Scenario 1:
key1 has two values v1 and v2. There're two snapshots s1 and s2 taken after v1 and v2 are committed. Right after compaction output v2, s1 is released. Now findEarliestVisibleSnapshot may see s1 being released, and return the next snapshot, which is s2. That's larger than v2's earliest visible snapshot, which was s1.
The fix: the only place we check against last snapshot and current key snapshot is when we decide whether to compact out a value if it is hidden by a later value. In the check if we see current snapshot is even larger than last snapshot, we know last snapshot is released, and we are safe to compact out current key.
Scenario 2:
key1 has two values v1 and v2. there are two snapshots s1 and s2 taken after v1 and v2 are committed. During compaction before we process the key, s1 is released. When compaction process v2, snapshot checker may return kSnapshotReleased, and the earliest visible snapshot for v2 become s2. When compaction process v1, snapshot checker may return kIsInSnapshot (for WritePrepared transaction, it could be because v1 is still in commit cache). The result will become inconsistent here.
The fix: remember the set of released snapshots ever reported by snapshot checker, and ignore them when finding result for findEarliestVisibleSnapshot.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4890
Differential Revision: D13705538
Pulled By: maysamyabandeh
fbshipit-source-id: e577f0d9ee1ff5a6035f26859e56902ecc85a5a4
Summary:
Compaction iterator keep a copy of list of live snapshots at the beginning of compaction, and then query snapshot checker to verify if values of a sequence number is visible to these snapshots. However when the snapshot is released in the middle of compaction, the snapshot checker implementation (i.e. WritePreparedSnapshotChecker) may remove info with the snapshot and may report incorrect result, which lead to values being compacted out when it shouldn't. This patch conservatively keep the values if snapshot checker determines that the snapshots is released.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4858
Differential Revision: D13617146
Pulled By: maysamyabandeh
fbshipit-source-id: cf18a94f6f61a94bcff73c280f117b224af5fbc3
Summary:
By convention, time_t almost always stores the integral number of seconds since
00:00 hours, Jan 1, 1970 UTC, according to http://www.cplusplus.com/reference/ctime/time_t/.
We surely want more precision than seconds.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4868
Differential Revision: D13633046
Pulled By: riversand963
fbshipit-source-id: 4e01e23a22e8838023c51a91247a286dbf3a5396
Summary:
With WritePrepared transaction, flush/compaction can contain uncommitted keys, and those keys can get committed during compaction. If a snapshot is taken before the key is committed, it should not see the key. On the other hand, compaction grab the list of snapshots at its beginning, and only consider those snapshots to dedup keys. Consider the case:
```
seq = 1: put "foo" = "bar"
seq = 2: transaction T: delete "foo", prepare
seq = 3: compaction start
seq = 4: take snapshot S
seq = 5: transaction T: commit.
...
seq = N: compaction iterator reached key "foo".
```
When compaction start, the list of snapshot is empty. Compaction doesn't take snapshot S into account. When it reached "foo", transaction T is committed. Compaction may think the value "foo=bar" is not visible by any snapshot (which is wrong), and compact the value out.
The fix is to explicitly take a snapshot before compaction grabbing the list of snapshots. Compaction will then has to keep keys visible to this snapshot.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4883
Differential Revision: D13668775
Pulled By: maysamyabandeh
fbshipit-source-id: 1cab9615f94b7d3e8522cc3d44c3a14c7d4720e4
Summary:
The AdvanceMaxEvictedSeq algorithm assumes that new snapshots always have sequence number larger than the last max_evicted_seq_. To enforce this assumption we make two changes:
i) max is not advanced beyond the last published seq, with the exception that the evicted commit entry itself is not published yet, which is quite rare.
ii) When obtaining the snapshot if the max_evicted_seq_ is not published yet, commit a dummy entry so that it waits for it to be published and also increased the latest published seq by one above the max.
To test these non-realistic corner cases we create a commit cache with size 1 so that every single commit results into eviction.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4886
Differential Revision: D13685270
Pulled By: maysamyabandeh
fbshipit-source-id: 5461bc09c2a9b75798bfcb9853a256c81cdac0b0
Summary:
Right now the error mesage when options.wal_dir doesn't exist is not helpful to users. Be more specific
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4874
Differential Revision: D13642425
Pulled By: siying
fbshipit-source-id: 9a3172ed0f799af233b0f3b2e5e35bc7ce04c7b5
Summary:
If we do not do this, then reading MutableCFOptions may have a race condition
with SetOptions which modifies MutableCFOptions.
Also reserve space in advance for vectors to avoid reallocation changing the
address of its elements.
Test plan
```
$make clean && make -j32 all check
$make clean && COMPILE_WITH_TSAN=1 make -j32 all check
$make clean && COMPILE_WITH_ASAN=1 make -j32 all check
```
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4876
Differential Revision: D13644500
Pulled By: riversand963
fbshipit-source-id: 4b8112c5c819d5a2922bb61ad1521b3d2fb2fd47
Summary:
The vector returned by SnapshotList::GetAll could have duplicate entries if two separate snapshots have the same sequence number. However, when this vector is used in compaction the duplicate entires are of no use and could be safely ignored. Moreover not having duplicate entires simplifies reasoning in the compaction_iterator.cc code. For example when searching for the previous_snap we currently use the snapshot before the current one but the way the code uses that it expects it to be also less than the current snapshot, which would be simpler to read if there is no duplicate entry in the snapshot list.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4860
Differential Revision: D13615502
Pulled By: maysamyabandeh
fbshipit-source-id: d45bf01213ead5f39db811f951802da6fcc3332b
Summary:
https://github.com/facebook/rocksdb/pull/3340 introduces preloading when max_open_files != -1.
It doesn't preload index and filter in non-initial file loading case. This is a little bit too
complicated to understand. We observed in one MyRocks use case where the filter is expected to be
preloaded but is not. To simplify the use case, we simply always prefetch the index and filter.
They anyway is expected to be loaded in the file verification phase anyway.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4852
Differential Revision: D13595402
Pulled By: siying
fbshipit-source-id: d4d8624eb3e849e20aeb990df2100502d85aff31
Summary:
In compaction iterator, if the next value of single delete is a blob value, it should not treated as mismatch. This is only a minor fix and doesn't affect correctness.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4848
Differential Revision: D13585812
Pulled By: yiwu-arbug
fbshipit-source-id: 0ff6223fa03a644ac9fd8a2d77f9d6711d0a62b0
Summary:
Previously for point lookup we decided which file to look into based on user key overlap only. We also did not truncate range tombstones in the point lookup code path. These two ideas did not interact well in cases like this:
- L1 has range tombstone [a, c)#1 and point key b#2. The data is split between file1 with range [a#1,1, b#72057594037927935,15], and file2 with range [b#2, c#1].
- L1's file2 gets compacted to L2.
- User issues `Get()` for b#3.
- L1's file1 is opened and the range tombstone [a, c)#1 is found for b, while no point-key for b is found in L1.
- `Get()` assumes that the range tombstone must cover all data in that range in lower levels, so short circuits and returns `NotFound`.
The solution to this problem is to not look into files that only overlap with the point lookup at a range tombstone sentinel endpoint. In the above example, this would mean not opening L1's file1 or its tombstones during the `Get()`.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4829
Differential Revision: D13561355
Pulled By: ajkr
fbshipit-source-id: a13c21c816870a2f5d32a48af6dbd719a7d9d19f
Summary:
as titled.
Since different bg flush threads can flush different sets of column families
(due to column family creation and drop), we decide not to let one thread
perform atomic flush result installation for other threads. Bg flush threads
will install their atomic flush results sequentially to MANIFEST, using
a conditional variable, i.e. atomic_flush_install_cv_ to coordinate.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4791
Differential Revision: D13498930
Pulled By: riversand963
fbshipit-source-id: dd7482fc41f4bd22dad1e1ef7d4764ef424688d7
Summary:
Declare Jemalloc non-standard APIs as weak symbols, so that if Jemalloc is linked with the binary, these symbols will be replaced by Jemalloc's, otherwise they will be nullptr. This is similar to how folly detect jemalloc, but we assume the main program use jemalloc as long as jemalloc is linked: https://github.com/facebook/folly/blob/master/folly/memory/Malloc.h#L147
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4844
Differential Revision: D13574934
Pulled By: yiwu-arbug
fbshipit-source-id: 7ea871beb1be7d5a1259cc38f9b78078793db2db
Summary:
The original implementation has two problems:
1. f0dda35d7d/db/db_impl_write.cc (L478)f0dda35d7d/db/write_thread.h (L231)
If the callback status of leader of the write_group fails, then the whole write_group will not write to WAL, this may cause data loss.
2. f0dda35d7d/db/write_thread.h (L130)
The annotation says that Writer.status is the status of memtable inserter, but the original implementation use it for another case which is not consistent with the original design. Looks like we can still reuse Writer.status, but we should modify the annotation, so Writer.status is not only the status of memtable inserter.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4838
Differential Revision: D13574070
Pulled By: yiwu-arbug
fbshipit-source-id: a2a2aefcfd329c4c6a91652bf090aaf1ce119c4b
Summary:
DBSSTTest.RateLimitedDelete is flakey. The root cause is not completely identified, but
the compaction waiting in the test doesn't strictly wait for compaction cleaning to finish, which
may cause test flakiness. Fix it first and see whether the failures still happen.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4840
Differential Revision: D13567273
Pulled By: siying
fbshipit-source-id: 6fce38b912aff92a925231e7aa9bb0fef892761a
Summary:
- To be consistent with the accounting of other optypes in `TableProperties`, we should count range tombstones in `TableProperties::num_entries` and `TableProperties::num_deletions`.
- Updated assertions in stress test's `OnTableFileCreated` handler to accept files with range tombstones only.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4841
Differential Revision: D13568424
Pulled By: ajkr
fbshipit-source-id: 0139d7806494eda20ece67ec460d2458dbbf6026
Summary:
Avoid locking the DB mutex in order to reference SuperVersions. Instead, we get the thread local cached SuperVersion for each column family in the list. It depends on finding a sequence number that overlaps with all the open memtables. We start with the latest published sequence number, and if any of the memtables is sealed before we can get all the SuperVersions, the process is repeated. After a few times, give up and lock the DB mutex.
Tests:
1. Unit tests
2. make check
3. db_bench -
TEST_TMPDIR=/dev/shm ./db_bench -use_existing_db=true -benchmarks=readrandom -write_buffer_size=4194304 -target_file_size_base=4194304 -max_bytes_for_level_base=16777216 -num=5000000 -reads=1000000 -threads=32 -compression_type=none -cache_size=1048576000 -batch_size=1 -bloom_bits=1
readrandom : 0.167 micros/op 5983920 ops/sec; 426.2 MB/s (1000000 of 1000000 found)
Multireadrandom with batch size 1:
multireadrandom : 0.176 micros/op 5684033 ops/sec; (1000000 of 1000000 found)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4754
Differential Revision: D13363550
Pulled By: anand1976
fbshipit-source-id: 6243e8de7dbd9c8bb490a8eca385da0c855b1dd4
Summary:
The `flush_reason` parameter in `DBImpl::InstallSuperVersionAndScheduleWork` is
not used. Remove it.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4816
Differential Revision: D13543218
Pulled By: riversand963
fbshipit-source-id: 8fc75d49462ce092e85aef0fe0c50936140db153
Summary:
Choose to preload some files if options.max_open_files != -1. This can slightly narrow the gap of performance between options.max_open_files is -1 and a large number. To avoid a significant regression to DB reopen speed if options.max_open_files != -1. Limit the files to preload in DB open time to 16.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/3340
Differential Revision: D6686945
Pulled By: siying
fbshipit-source-id: 8ec11bbdb46e3d0cdee7b6ad5897a09c5a07869f
Summary:
1. Remove unused API SubtractCompactionTask().
2. Assert outstanding tasks drop to zero in ConcurrentTaskLimiterImpl destructor.
3. Remove GetOutstandingTask() check from manual compaction test, as TEST_WaitForCompact() doesn't synced with 'delete prepicked_compaction' in DBImpl::BGWorkCompaction(), which may make the test flaky.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4795
Differential Revision: D13542183
Pulled By: siying
fbshipit-source-id: 5eb2a47e62efe4126937149aa0df6e243ebefc33
