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:
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:
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:
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:
Now that v2 is fully functional, the v1 aggregator is removed.
The v2 aggregator has been renamed.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4778
Differential Revision: D13495930
Pulled By: abhimadan
fbshipit-source-id: 9d69500a60a283e79b6c4fa938fc68a8aa4d40d6
Summary:
The old RangeDelAggregator did expensive pre-processing work
to create a collapsed, binary-searchable representation of range
tombstones. With FragmentedRangeTombstoneIterator, much of this work is
now unnecessary. RangeDelAggregatorV2 takes advantage of this by seeking
in each iterator to find a covering tombstone in ShouldDelete, while
doing minimal work in AddTombstones. The old RangeDelAggregator is still
used during flush/compaction for now, though RangeDelAggregatorV2 will
support those uses in a future PR.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4649
Differential Revision: D13146964
Pulled By: abhimadan
fbshipit-source-id: be29a4c020fc440500c137216fcc1cf529571eb3
Summary:
Ran the following commands to recursively change all the files under RocksDB:
```
find . -type f -name "*.cc" -exec sed -i 's/ unique_ptr/ std::unique_ptr/g' {} +
find . -type f -name "*.cc" -exec sed -i 's/<unique_ptr/<std::unique_ptr/g' {} +
find . -type f -name "*.cc" -exec sed -i 's/ shared_ptr/ std::shared_ptr/g' {} +
find . -type f -name "*.cc" -exec sed -i 's/<shared_ptr/<std::shared_ptr/g' {} +
```
Running `make format` updated some formatting on the files touched.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4638
Differential Revision: D12934992
Pulled By: sagar0
fbshipit-source-id: 45a15d23c230cdd64c08f9c0243e5183934338a8
Summary:
Previously, range tombstones were accumulated from every level, which
was necessary if a range tombstone in a higher level covered a key in a lower
level. However, RangeDelAggregator::AddTombstones's complexity is based on
the number of tombstones that are currently stored in it, which is wasteful in
the Get case, where we only need to know the highest sequence number of range
tombstones that cover the key from higher levels, and compute the highest covering
sequence number at the current level. This change introduces this optimization, and
removes the use of RangeDelAggregator from the Get path.
In the benchmark results, the following command was used to initialize the database:
```
./db_bench -db=/dev/shm/5k-rts -use_existing_db=false -benchmarks=filluniquerandom -write_buffer_size=1048576 -compression_type=lz4 -target_file_size_base=1048576 -max_bytes_for_level_base=4194304 -value_size=112 -key_size=16 -block_size=4096 -level_compaction_dynamic_level_bytes=true -num=5000000 -max_background_jobs=12 -benchmark_write_rate_limit=20971520 -range_tombstone_width=100 -writes_per_range_tombstone=100 -max_num_range_tombstones=50000 -bloom_bits=8
```
...and the following command was used to measure read throughput:
```
./db_bench -db=/dev/shm/5k-rts/ -use_existing_db=true -benchmarks=readrandom -disable_auto_compactions=true -num=5000000 -reads=100000 -threads=32
```
The filluniquerandom command was only run once, and the resulting database was used
to measure read performance before and after the PR. Both binaries were compiled with
`DEBUG_LEVEL=0`.
Readrandom results before PR:
```
readrandom : 4.544 micros/op 220090 ops/sec; 16.9 MB/s (63103 of 100000 found)
```
Readrandom results after PR:
```
readrandom : 11.147 micros/op 89707 ops/sec; 6.9 MB/s (63103 of 100000 found)
```
So it's actually slower right now, but this PR paves the way for future optimizations (see #4493).
----
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4449
Differential Revision: D10370575
Pulled By: abhimadan
fbshipit-source-id: 9a2e152be1ef36969055c0e9eb4beb0d96c11f4d
Summary:
There was a bug that the user comparator would receive the internal key instead of the user key. The bug was due to RangeMightExistAfterSortedRun expecting user key but receiving internal key when called in GenerateBottommostFiles. The patch augment an existing unit test to reproduce the bug and fixes it.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4575
Differential Revision: D10500434
Pulled By: maysamyabandeh
fbshipit-source-id: 858346d2fd102cce9e20516d77338c112bdfe366
Summary:
Level compaction usually performs poorly when the writes so heavy that the level targets can't be guaranteed. With this improvement, we improve level_compaction_dynamic_level_bytes = true so that in the write heavy cases, the level multiplier can be slightly adjusted based on the size of L0.
We keep the behavior the same if number of L0 files is under 2X compaction trigger and the total size is less than options.max_bytes_for_level_base, so that unless write is so heavy that compaction cannot keep up, the behavior doesn't change.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4338
Differential Revision: D9636782
Pulled By: siying
fbshipit-source-id: e27fc17a7c29c84b00064cc17536a01dacef7595
Summary:
When a CompactRange() call for a level is truncated before the end key
is reached, because it exceeds max_compaction_bytes, we need to properly
set the compaction_end parameter to indicate the stop key. The next
CompactRange will use that as the begin key. We set it to the smallest
key of the next file in the level after expanding inputs to get a clean
cut.
Previously, we were setting it before expanding inputs. So we could end
up recompacting some files. In a pathological case, where a single key
has many entries spanning all the files in the level (possibly due to
merge operands without a partial merge operator, thus resulting in
compaction output identical to the input), this would result in
an endless loop over the same set of files.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4496
Differential Revision: D10395026
Pulled By: anand1976
fbshipit-source-id: f0c2f89fee29b4b3be53b6467b53abba8e9146a9
Summary:
Leverage existing `FlushJob` to implement atomic flush of multiple column families.
This PR depends on other PRs and is a subset of #3752 . This PR itself is not sufficient in fulfilling atomic flush.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4262
Differential Revision: D9283109
Pulled By: riversand963
fbshipit-source-id: 65401f913e4160b0a61c0be6cd02adc15dad28ed
Summary:
During recovery, RocksDB is able to handle version edits that belong to group commits.
This PR is a subset of [PR 3752](https://github.com/facebook/rocksdb/pull/3752)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/3945
Differential Revision: D8529122
Pulled By: riversand963
fbshipit-source-id: 57cb0f9cc55ecca684a837742d6626dc9c07f37e
Summary:
Given that index value is a BlockHandle, which is basically an <offset, size> pair we can apply delta encoding on the values. The first value at each index restart interval encoded the full BlockHandle but the rest encode only the size. Refer to IndexBlockIter::DecodeCurrentValue for the detail of the encoding. This reduces the index size which helps using the block cache more efficiently. The feature is enabled with using format_version 4.
The feature comes with a bit of cpu overhead which should be paid back by the higher cache hits due to smaller index block size.
Results with sysbench read-only using 4k blocks and using 16 index restart interval:
Format 2:
19585 rocksdb read-only range=100
Format 3:
19569 rocksdb read-only range=100
Format 4:
19352 rocksdb read-only range=100
Pull Request resolved: https://github.com/facebook/rocksdb/pull/3983
Differential Revision: D8361343
Pulled By: maysamyabandeh
fbshipit-source-id: f882ee082322acac32b0072e2bdbb0b5f854e651
Summary:
If crash happen after a hard link established, Recover function may reuse the file number that has already assigned to the internal file, and this will overwrite the external file. To protect the external file, we have to make sure the file number will never being reused.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4099
Differential Revision: D9034092
Pulled By: riversand963
fbshipit-source-id: 3f1a737440b86aa2ef01673e5013aacbb7c33e28
Summary:
Do not consider the range tombstone sentinel key as causing 2 adjacent
sstables in a level to overlap. When a range tombstone's end key is the
largest key in an sstable, the sstable's end key is so to a "sentinel"
value that is the smallest key in the next sstable with a sequence
number of kMaxSequenceNumber. This "sentinel" is guaranteed to not
overlap in internal-key space with the next sstable. Unfortunately,
GetOverlappingFiles uses user-keys to determine overlap and was thus
considering 2 adjacent sstables in a level to overlap if they were
separated by this sentinel key. This in turn would cause compactions to
be larger than necessary.
Note that this conflicts with
https://github.com/facebook/rocksdb/pull/2769 and cases
`DBRangeDelTest.CompactionTreatsSplitInputLevelDeletionAtomically` to
fail.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4050
Differential Revision: D8844423
Pulled By: ajkr
fbshipit-source-id: df3f9f1db8f4cff2bff77376b98b83c2ae1d155b
Summary:
This PR supports the group commit of multiple version edit entries corresponding to different column families. Column family drop/creation still cannot be grouped. This PR is a subset of [PR 3752](https://github.com/facebook/rocksdb/pull/3752).
Closes https://github.com/facebook/rocksdb/pull/3944
Differential Revision: D8432536
Pulled By: riversand963
fbshipit-source-id: 8f11bd05193b6c0d9272d82e44b676abfac113cb
Summary:
Currently it is not possible to change bloom filter config without restart the db, which is causing a lot of operational complexity for users.
This PR aims to make it possible to dynamically change bloom filter config.
Closes https://github.com/facebook/rocksdb/pull/3601
Differential Revision: D7253114
Pulled By: miasantreble
fbshipit-source-id: f22595437d3e0b86c95918c484502de2ceca120c
Summary:
This patch record min log number to keep to the manifest while flushing SST files to ignore them and any WAL older than them during recovery. This is to avoid scenarios when we have a gap between the WAL files are fed to the recovery procedure. The gap could happen by for example out-of-order WAL deletion. Such gap could cause problems in 2PC recovery where the prepared and commit entry are placed into two separate WAL and gap in the WALs could result into not processing the WAL with the commit entry and hence breaking the 2PC recovery logic.
Before the commit, for 2PC case, we determined which log number to keep in FindObsoleteFiles(). We looked at the earliest logs with outstanding prepare entries, or prepare entries whose respective commit or abort are in memtable. With the commit, the same calculation is done while we apply the SST flush. Just before installing the flush file, we precompute the earliest log file to keep after the flush finishes using the same logic (but skipping the memtables just flushed), record this information to the manifest entry for this new flushed SST file. This pre-computed value is also remembered in memory, and will later be used to determine whether a log file can be deleted. This value is unlikely to change until next flush because the commit entry will stay in memtable. (In WritePrepared, we could have removed the older log files as soon as all prepared entries are committed. It's not yet done anyway. Even if we do it, the only thing we loss with this new approach is earlier log deletion between two flushes, which does not guarantee to happen anyway because the obsolete file clean-up function is only executed after flush or compaction)
This min log number to keep is stored in the manifest using the safely-ignore customized field of AddFile entry, in order to guarantee that the DB generated using newer release can be opened by previous releases no older than 4.2.
Closes https://github.com/facebook/rocksdb/pull/3765
Differential Revision: D7747618
Pulled By: siying
fbshipit-source-id: d00c92105b4f83852e9754a1b70d6b64cb590729
Summary:
This reverts commit 73f21a7b21.
It breaks compatibility. When created a DB using a build with this new change, opening the DB and reading the data will fail with this error:
"Corruption: Can't access /000000.sst: IO error: while stat a file for size: /tmp/xxxx/000000.sst: No such file or directory"
This is because the dummy AddFile4 entry generated by the new code will be treated as a real entry by an older build. The older build will think there is a real file with number 0, but there isn't such a file.
Closes https://github.com/facebook/rocksdb/pull/3762
Differential Revision: D7730035
Pulled By: siying
fbshipit-source-id: f2051859eff20ef1837575ecb1e1bb96b3751e77
Summary:
In this change, an option to set different paths for different column families is added.
This option is set via cf_paths setting of ColumnFamilyOptions. This option will work in a similar fashion to db_paths setting. Cf_paths is a vector of Dbpath values which contains a pair of the absolute path and target size. Multiple levels in a Column family can go to different paths if cf_paths has more than one path.
To maintain backward compatibility, if cf_paths is not specified for a column family, db_paths setting will be used. Note that, if db_paths setting is also not specified, RocksDB already has code to use db_name as the only path.
Changes :
1) A new member "cf_paths" is added to ImmutableCfOptions. This is set, based on cf_paths setting of ColumnFamilyOptions and db_paths setting of ImmutableDbOptions. This member is used to identify the path information whenever files are accessed.
2) Validation checks are added for cf_paths setting based on existing checks for db_paths setting.
3) DestroyDB, PurgeObsoleteFiles etc. are edited to support multiple cf_paths.
4) Unit tests are added appropriately.
Closes https://github.com/facebook/rocksdb/pull/3102
Differential Revision: D6951697
Pulled By: ajkr
fbshipit-source-id: 60d2262862b0a8fd6605b09ccb0da32bb331787d
Summary:
Level Compaction with TTL.
As of today, a file could exist in the LSM tree without going through the compaction process for a really long time if there are no updates to the data in the file's key range. For example, in certain use cases, the keys are not actually "deleted"; instead they are just set to empty values. There might not be any more writes to this "deleted" key range, and if so, such data could remain in the LSM for a really long time resulting in wasted space.
Introducing a TTL could solve this problem. Files (and, in turn, data) older than TTL will be scheduled for compaction when there is no other background work. This will make the data go through the regular compaction process and get rid of old unwanted data.
This also has the (good) side-effect of all the data in the non-bottommost level being newer than ttl, and all data in the bottommost level older than ttl. It could lead to more writes while reducing space.
This functionality can be controlled by the newly introduced column family option -- ttl.
TODO for later:
- Make ttl mutable
- Extend TTL to Universal compaction as well? (TTL is already supported in FIFO)
- Maybe deprecate CompactionOptionsFIFO.ttl in favor of this new ttl option.
Closes https://github.com/facebook/rocksdb/pull/3591
Differential Revision: D7275442
Pulled By: sagar0
fbshipit-source-id: dcba484717341200d419b0953dafcdf9eb2f0267
Summary:
This patch record the deleted WAL numbers in the manifest to ignore them and any WAL older than them during recovery. This is to avoid scenarios when we have a gap between the WAL files are fed to the recovery procedure. The gap could happen by for example out-of-order WAL deletion. Such gap could cause problems in 2PC recovery where the prepared and commit entry are placed into two separate WAL and gap in the WALs could result into not processing the WAL with the commit entry and hence breaking the 2PC recovery logic.
Closes https://github.com/facebook/rocksdb/pull/3488
Differential Revision: D6967893
Pulled By: maysamyabandeh
fbshipit-source-id: 13119feb155a08ab6d4909f437c7a750480dc8a1
Summary:
If there are a lot of overlapped files in L0, creating a merging iterator for
all files in L0 to check overlap can be very slow because we need to read and
seek all files in L0. However, in that case, the ingested file is likely to
overlap with some files in L0, so if we check those files one by one, we can stop
once we encounter overlap.
Ref: https://github.com/facebook/rocksdb/issues/3540
Closes https://github.com/facebook/rocksdb/pull/3564
Differential Revision: D7196784
Pulled By: anand1976
fbshipit-source-id: 8700c1e903bd515d0fa7005b6ce9b3a3d9db2d67
Summary:
Add "rocksdb.live-sst-files-size" DB property which only include files of latest version. Existing "rocksdb.total-sst-files-size" include files from all versions and thus include files that's obsolete but not yet deleted. I'm going to use this new property to cap blob db sst + blob files size.
Closes https://github.com/facebook/rocksdb/pull/3548
Differential Revision: D7116939
Pulled By: yiwu-arbug
fbshipit-source-id: c6a52e45ce0f24ef78708156e1a923c1dd6bc79a
Summary:
Add PreReleaseCallback to be called at the end of WriteImpl but before publishing the sequence number. The callback is used in WritePrepareTxn to i) update the commit map, ii) update the last published sequence number in the 2nd write queue. It also ensures that all the commits will go to the 2nd queue.
These changes will ensure that the commit map is updated before the sequence number is published and used by reading snapshots. If we use two write queues, the snapshots will use the seq number published by the 2nd queue. If we use one write queue (the default, the snapshots will use the last seq number in the memtable, which also indicates the last published seq number.
Closes https://github.com/facebook/rocksdb/pull/3205
Differential Revision: D6438959
Pulled By: maysamyabandeh
fbshipit-source-id: f8b6c434e94bc5f5ab9cb696879d4c23e2577ab9
Summary:
Before we were checking every file in the level which was unnecessary. We can piggyback onto the code for checking point-key overlap, which already opens all the files that could possibly contain overlapping range deletions. This PR makes us check just the range deletions from those files, so no extra ones will be opened.
Closes https://github.com/facebook/rocksdb/pull/3179
Differential Revision: D6358125
Pulled By: ajkr
fbshipit-source-id: 00e200770fdb8f3cc6b1b2da232b755e4ba36279
Summary:
When snapshots are held for a long time, files may reach the bottom level containing overwritten/deleted keys. We previously had no mechanism to trigger compaction on such files. This particularly impacted DBs that write to different parts of the keyspace over time, as such files would never be naturally compacted due to second-last level files moving down. This PR introduces a mechanism for bottommost files to be recompacted upon releasing all snapshots that prevent them from dropping their deleted/overwritten keys.
- Changed `CompactionPicker` to compact files in `BottommostFilesMarkedForCompaction()`. These are the last choice when picking. Each file will be compacted alone and output to the same level in which it originated. The goal of this type of compaction is to rewrite the data excluding deleted/overwritten keys.
- Changed `ReleaseSnapshot()` to recompute the bottom files marked for compaction when the oldest existing snapshot changes, and schedule a compaction if needed. We cache the value that oldest existing snapshot needs to exceed in order for another file to be marked in `bottommost_files_mark_threshold_`, which allows us to avoid recomputing marked files for most snapshot releases.
- Changed `VersionStorageInfo` to track the list of bottommost files, which is recomputed every time the version changes by `UpdateBottommostFiles()`. The list of marked bottommost files is first computed in `ComputeBottommostFilesMarkedForCompaction()` when the version changes, but may also be recomputed when `ReleaseSnapshot()` is called.
- Extracted core logic of `Compaction::IsBottommostLevel()` into `VersionStorageInfo::RangeMightExistAfterSortedRun()` since logic to check whether a file is bottommost is now necessary outside of compaction.
Closes https://github.com/facebook/rocksdb/pull/3009
Differential Revision: D6062044
Pulled By: ajkr
fbshipit-source-id: 123d201cf140715a7d5928e8b3cb4f9cd9f7ad21
Summary:
The file numbers assigned post-repair were sometimes smaller than older files' numbers due to `LogAndApply` saving the wrong next file number in the manifest.
- Mark the highest file seen during repair as used before `LogAndApply` so the correct next file number will be stored.
- Renamed `MarkFileNumberUsedDuringRecovery` to `MarkFileNumberUsed` since now it's used during repair in addition to during recovery
- Added `TEST_Current_Next_FileNo` to expose the next file number for the unit test.
Closes https://github.com/facebook/rocksdb/pull/2988
Differential Revision: D6018083
Pulled By: ajkr
fbshipit-source-id: 3f25cbf74439cb8f16dd12af90b67f9f9f75e718
Summary:
Add kTypeBlobIndex value type, which will be used by blob db only, to insert a (key, blob_offset) KV pair. The purpose is to
1. Make it possible to open existing rocksdb instance as blob db. Existing value will be of kTypeIndex type, while value inserted by blob db will be of kTypeBlobIndex.
2. Make rocksdb able to detect if the db contains value written by blob db, if so return error.
3. Make it possible to have blob db optionally store value in SST file (with kTypeValue type) or as a blob value (with kTypeBlobIndex type).
The root db (DBImpl) basically pretended kTypeBlobIndex are normal value on write. On Get if is_blob is provided, return whether the value read is of kTypeBlobIndex type, or return Status::NotSupported() status if is_blob is not provided. On scan allow_blob flag is pass and if the flag is true, return wether the value is of kTypeBlobIndex type via iter->IsBlob().
Changes on blob db side will be in a separate patch.
Closes https://github.com/facebook/rocksdb/pull/2886
Differential Revision: D5838431
Pulled By: yiwu-arbug
fbshipit-source-id: 3c5306c62bc13bb11abc03422ec5cbcea1203cca
Summary:
This patch instruments the read path to verify each read value against an optional ReadCallback class. If the value is rejected, the reader moves on to the next value. The WritePreparedTxn makes use of this feature to skip sequence numbers that are not in the read snapshot.
Closes https://github.com/facebook/rocksdb/pull/2850
Differential Revision: D5787375
Pulled By: maysamyabandeh
fbshipit-source-id: 49d808b3062ab35e7ae98ad388f659757794184c
Summary:
When we had a single thread pool for compactions, a thread could be busy for a long time (minutes) executing a compaction involving the bottom level. In multi-instance setups, the entire thread pool could be consumed by such bottom-level compactions. Then, top-level compactions (e.g., a few L0 files) would be blocked for a long time ("head-of-line blocking"). Such top-level compactions are critical to prevent compaction stalls as they can quickly reduce number of L0 files / sorted runs.
This diff introduces a bottom-priority queue for universal compactions including the bottom level. This alleviates the head-of-line blocking situation for fast, top-level compactions.
- Added `Env::Priority::BOTTOM` thread pool. This feature is only enabled if user explicitly configures it to have a positive number of threads.
- Changed `ThreadPoolImpl`'s default thread limit from one to zero. This change is invisible to users as we call `IncBackgroundThreadsIfNeeded` on the low-pri/high-pri pools during `DB::Open` with values of at least one. It is necessary, though, for bottom-pri to start with zero threads so the feature is disabled by default.
- Separated `ManualCompaction` into two parts in `PrepickedCompaction`. `PrepickedCompaction` is used for any compaction that's picked outside of its execution thread, either manual or automatic.
- Forward universal compactions involving last level to the bottom pool (worker thread's entry point is `BGWorkBottomCompaction`).
- Track `bg_bottom_compaction_scheduled_` so we can wait for bottom-level compactions to finish. We don't count them against the background jobs limits. So users of this feature will get an extra compaction for free.
Closes https://github.com/facebook/rocksdb/pull/2580
Differential Revision: D5422916
Pulled By: ajkr
fbshipit-source-id: a74bd11f1ea4933df3739b16808bb21fcd512333
Summary:
This reverts the previous commit 1d7048c598, which broke the build.
Did a `git revert 1d7048c`.
Closes https://github.com/facebook/rocksdb/pull/2627
Differential Revision: D5476473
Pulled By: sagar0
fbshipit-source-id: 4756ff5c0dfc88c17eceb00e02c36176de728d06
Summary: This uses `clang-tidy` to comment out unused parameters (in functions, methods and lambdas) in fbcode. Cases that the tool failed to handle are fixed manually.
Reviewed By: igorsugak
Differential Revision: D5454343
fbshipit-source-id: 5dee339b4334e25e963891b519a5aa81fbf627b2
Summary:
Throughput: 46k tps in our sysbench settings (filling the details later)
The idea is to have the simplest change that gives us a reasonable boost
in 2PC throughput.
Major design changes:
1. The WAL file internal buffer is not flushed after each write. Instead
it is flushed before critical operations (WAL copy via fs) or when
FlushWAL is called by MySQL. Flushing the WAL buffer is also protected
via mutex_.
2. Use two sequence numbers: last seq, and last seq for write. Last seq
is the last visible sequence number for reads. Last seq for write is the
next sequence number that should be used to write to WAL/memtable. This
allows to have a memtable write be in parallel to WAL writes.
3. BatchGroup is not used for writes. This means that we can have
parallel writers which changes a major assumption in the code base. To
accommodate for that i) allow only 1 WriteImpl that intends to write to
memtable via mem_mutex_--which is fine since in 2PC almost all of the memtable writes
come via group commit phase which is serial anyway, ii) make all the
parts in the code base that assumed to be the only writer (via
EnterUnbatched) to also acquire mem_mutex_, iii) stat updates are
protected via a stat_mutex_.
Note: the first commit has the approach figured out but is not clean.
Submitting the PR anyway to get the early feedback on the approach. If
we are ok with the approach I will go ahead with this updates:
0) Rebase with Yi's pipelining changes
1) Currently batching is disabled by default to make sure that it will be
consistent with all unit tests. Will make this optional via a config.
2) A couple of unit tests are disabled. They need to be updated with the
serial commit of 2PC taken into account.
3) Replacing BatchGroup with mem_mutex_ got a bit ugly as it requires
releasing mutex_ beforehand (the same way EnterUnbatched does). This
needs to be cleaned up.
Closes https://github.com/facebook/rocksdb/pull/2345
Differential Revision: D5210732
Pulled By: maysamyabandeh
fbshipit-source-id: 78653bd95a35cd1e831e555e0e57bdfd695355a4
Summary:
We estimate number of reads per SST files, by updating the counter per file in sampled read requests. This information can later be used to trigger compactions to improve read performacne.
Closes https://github.com/facebook/rocksdb/pull/2417
Differential Revision: D5193528
Pulled By: siying
fbshipit-source-id: b4241c5ad0eaf444b61afb53f8e6290d9f5da2df
Summary:
The range deletion meta-block iterators weren't getting cleaned up properly since they don't support arena allocation. I didn't implement arena support since, in the general case, each iterator is used only once and separately from all other iterators, so there should be no benefit to data locality.
Anyways, this diff fixes up #2370 by treating range deletion iterators as non-arena-allocated.
Closes https://github.com/facebook/rocksdb/pull/2399
Differential Revision: D5171119
Pulled By: ajkr
fbshipit-source-id: bef6f5c4c5905a124f4993945aed4bd86e2807d8
Summary:
Previously we returned NotSupported when ingesting files into a database containing any range deletions. This diff adds the support.
- Flush if any memtable contains range deletions overlapping the to-be-ingested file
- Place to-be-ingested file before any level that contains range deletions overlapping it.
- Added support for `Version` to return iterators over range deletions in a given level. Previously, we piggybacked getting range deletions onto `Version`'s `Get()` / `AddIterator()` functions by passing them a `RangeDelAggregator*`. But file ingestion needs to get iterators over range deletions, not populate an aggregator (since the aggregator does collapsing and doesn't expose the actual ranges).
Closes https://github.com/facebook/rocksdb/pull/2370
Differential Revision: D5127648
Pulled By: ajkr
fbshipit-source-id: 816faeb9708adfa5287962bafdde717db56e3f1a
Summary:
Move some files under util/ to new directories env/, monitoring/ options/ and cache/
Closes https://github.com/facebook/rocksdb/pull/2090
Differential Revision: D4833681
Pulled By: siying
fbshipit-source-id: 2fd8bef
Summary:
PinnableSlice
Summary:
Currently the point lookup values are copied to a string provided by the
user. This incures an extra memcpy cost. This patch allows doing point lookup
via a PinnableSlice which pins the source memory location (instead of
copying their content) and releases them after the content is consumed
by the user. The old API of Get(string) is translated to the new API
underneath.
Here is the summary for improvements:
value 100 byte: 1.8% regular, 1.2% merge values
value 1k byte: 11.5% regular, 7.5% merge values
value 10k byte: 26% regular, 29.9% merge values
The improvement for merge could be more if we extend this approach to
pin the merge output and delay the full merge operation until the user
actually needs it. We have put that for future work.
PS:
Sometimes we observe a small decrease in performance when switching from
t5452014 to this patch but with the old Get(string) API. The d
Closes https://github.com/facebook/rocksdb/pull/1756
Differential Revision: D4391738
Pulled By: maysamyabandeh
fbshipit-source-id: 6f3edd3
