Summary:
This is implemented by extending ReadCallback with another function `MaxUnpreparedSequenceNumber` which returns the largest visible sequence number for the current transaction, if there is uncommitted data written to DB. Otherwise, it returns zero, indicating no uncommitted data.
There are the places where reads had to be modified.
- Get and Seek/Next was just updated to seek to max(snapshot_seq, MaxUnpreparedSequenceNumber()) instead, and iterate until a key was visible.
- Prev did not need need updates since it did not use the Seek to sequence number optimization. Assuming that locks were held when writing unprepared keys, and ValidateSnapshot runs, there should only be committed keys and unprepared keys of the current transaction, all of which are visible. Prev will simply iterate to get the last visible key.
- Reseeking to skip keys optimization was also disabled for write unprepared, since it's possible to hit the max_skip condition even while reseeking. There needs to be some way to resolve infinite looping in this case.
Closes https://github.com/facebook/rocksdb/pull/3955
Differential Revision: D8286688
Pulled By: lth
fbshipit-source-id: 25e42f47fdeb5f7accea0f4fd350ef35198caafe
Summary:
Add a new table property, rocksdb.num.range-deletions, which tracks the
number of range deletions in a block-based table. Range deletions are no
longer counted in rocksdb.num.entries; as discovered in PR #3778, there
are various code paths that implicitly assume that rocksdb.num.entries
counts only true keys, not range deletions.
/cc ajkr nvanbenschoten
Closes https://github.com/facebook/rocksdb/pull/4016
Differential Revision: D8527575
Pulled By: ajkr
fbshipit-source-id: 92e7edbe78fda53756a558013c9fb496e7764fd7
Summary:
Previously in https://github.com/facebook/rocksdb/pull/3601 bloom filter will only be checked if `prefix_extractor` in the mutable_cf_options matches the one found in the SST file.
This PR relaxes the requirement by checking if all keys in the range [user_key, iterate_upper_bound) all share the same prefix after transforming using the BF in the SST file. If so, the bloom filter is considered compatible and will continue to be looked at.
Closes https://github.com/facebook/rocksdb/pull/3899
Differential Revision: D8157459
Pulled By: miasantreble
fbshipit-source-id: 18d17cba56a1005162f8d5db7a27aba277089c41
Summary:
Pass in `for_compaction` to `BlockBasedTableIterator` via `BlockBasedTableReader::NewIterator`.
In 7103559f49, `for_compaction` was set in `BlockBasedTable::Rep` via `BlockBasedTable::SetupForCompaction`. In hindsight it was not the right decision; it also caused TSAN to complain.
Closes https://github.com/facebook/rocksdb/pull/4048
Differential Revision: D8601056
Pulled By: sagar0
fbshipit-source-id: 30127e898c15c38c1080d57710b8c5a6d64a0ab3
Summary:
Top-level index in partitioned index/filter blocks are small and could be pinned in memory. So far we use that by cache_index_and_filter_blocks to false. This however make it difficult to keep account of the total memory usage. This patch introduces pin_top_level_index_and_filter which in combination with cache_index_and_filter_blocks=true keeps the top-level index in cache and yet pinned them to avoid cache misses and also cache lookup overhead.
Closes https://github.com/facebook/rocksdb/pull/4037
Differential Revision: D8596218
Pulled By: maysamyabandeh
fbshipit-source-id: 3a5f7f9ca6b4b525b03ff6bd82354881ae974ad2
Summary:
This PR extends the improvements in #3282 to also work when using Direct IO.
We see **4.5X performance improvement** in seekrandom benchmark doing long range scans, when using direct reads, on flash.
**Description:**
This change improves the performance of iterators doing long range scans (e.g. big/full index or table scans in MyRocks) by using readahead and prefetching additional data on each disk IO, and storing in a local buffer. This prefetching is automatically enabled on noticing more than 2 IOs for the same table file during iteration. The readahead size starts with 8KB and is exponentially increased on each additional sequential IO, up to a max of 256 KB. This helps in cutting down the number of IOs needed to complete the range scan.
**Implementation Details:**
- Used `FilePrefetchBuffer` as the underlying buffer to store the readahead data. `FilePrefetchBuffer` can now take file_reader, readahead_size and max_readahead_size as input to the constructor, and automatically do readahead.
- `FilePrefetchBuffer::TryReadFromCache` can now call `FilePrefetchBuffer::Prefetch` if readahead is enabled.
- `AlignedBuffer` (which is the underlying store for `FilePrefetchBuffer`) now takes a few additional args in `AlignedBuffer::AllocateNewBuffer` to allow copying data from the old buffer.
- Made sure not to re-read partial chunks of data that were already available in the buffer, from device again.
- Fixed a couple of cases where `AlignedBuffer::cursize_` was not being properly kept up-to-date.
**Constraints:**
- Similar to #3282, this gets currently enabled only when ReadOptions.readahead_size = 0 (which is the default value).
- Since the prefetched data is stored in a temporary buffer allocated on heap, this could increase the memory usage if you have many iterators doing long range scans simultaneously.
- Enabled only for user reads, and disabled for compactions. Compaction reads are controlled by the options `use_direct_io_for_flush_and_compaction` and `compaction_readahead_size`, and the current feature takes precautions not to mess with them.
**Benchmarks:**
I used the same benchmark as used in #3282.
Data fill:
```
TEST_TMPDIR=/data/users/$USER/benchmarks/iter ./db_bench -benchmarks=fillrandom -num=1000000000 -compression_type="none" -level_compaction_dynamic_level_bytes
```
Do a long range scan: Seekrandom with large number of nexts
```
TEST_TMPDIR=/data/users/$USER/benchmarks/iter ./db_bench -benchmarks=seekrandom -use_direct_reads -duration=60 -num=1000000000 -use_existing_db -seek_nexts=10000 -statistics -histogram
```
```
Before:
seekrandom : 37939.906 micros/op 26 ops/sec; 29.2 MB/s (1636 of 1999 found)
With this change:
seekrandom : 8527.720 micros/op 117 ops/sec; 129.7 MB/s (6530 of 7999 found)
```
~4.5X perf improvement. Taken on an average of 3 runs.
Closes https://github.com/facebook/rocksdb/pull/3884
Differential Revision: D8082143
Pulled By: sagar0
fbshipit-source-id: 4d7a8561cbac03478663713df4d31ad2620253bb
Summary:
https://github.com/facebook/rocksdb/pull/3764 introduced an optimization feature to skip duplicate prefix entires in full bloom filters. Unfortunately it also introduces a bug in partitioned full filters, where the duplicate prefix should still be inserted if it is in a new partition. The patch fixes the bug by resetting the duplicate detection logic each time a partition is cut.
This bug could result into false negatives, which means that DB could skip an existing key.
Closes https://github.com/facebook/rocksdb/pull/4024
Differential Revision: D8518866
Pulled By: maysamyabandeh
fbshipit-source-id: 044f4d988e606a330ecafd8c79daceb68b8796bf
Summary:
b555ed30a4 makes the BlockBasedTableIterator to be invalidated if the current position if over the upper bound. However, this can bring performance regression to the case of multiple Seek()s hitting the same data block but all out of upper bound.
For example, if an SST file has a data block containing following keys : {a, z}
The user sets the upper bound to be "x", and it executed following queries:
Seek("b")
Seek("c")
Seek("d")
Before the upper bound optimization, these queries always come to this same current data block of the iterator, but now inside each Seek() the data block is read from the block cache but is returned again.
To prevent this regression case, we keep the current data block iterator if it is upper bound.
Closes https://github.com/facebook/rocksdb/pull/4004
Differential Revision: D8463192
Pulled By: siying
fbshipit-source-id: 8710628b30acde7063a097c3184d6c4333a8ef81
Summary:
The Block object assumes contents are uncompressed. Block's constructor tries to read the number of restarts, but does not get an accurate number when its contents are compressed, which is causing issues like https://github.com/facebook/rocksdb/issues/3843.
This PR address this issue by skipping reconstruction of restart points when blocks are known to be compressed. Somehow the restart points can be read directly when Snappy is used and some tests (for example https://github.com/facebook/rocksdb/blob/master/db/db_block_cache_test.cc#L196) expects blocks to be fully constructed even when Snappy compression is used, so here we keep the restart point logic for Snappy.
Closes https://github.com/facebook/rocksdb/pull/3996
Differential Revision: D8416186
Pulled By: miasantreble
fbshipit-source-id: 002c0b62b9e5d89fb7736563d354ce0023c8cb28
Summary:
A recent change pushed down the upper bound checking to child iterators. However, this causes the logic of following sequence wrong:
Seek(key);
if (!Valid()) SeekToLast();
Because !Valid() may be caused by upper bounds, rather than the end of the iterator. In this case SeekToLast() points to totally wrong places. This can cause wrong results, infinite loops, or segfault in some cases.
This sequence is called when changing direction from forward to backward. And this by itself also implicitly happen during reseeking optimization in Prev().
Fix this bug by using SeekForPrev() rather than this sequuence, as what is already done in prefix extrator case.
Closes https://github.com/facebook/rocksdb/pull/3989
Differential Revision: D8385422
Pulled By: siying
fbshipit-source-id: 429e869990cfd2dc389421e0836fc496bed67bb4
Summary:
The sixth argument should be `key_includes_seq` bool, the seventh a `GetContext*`. We were mistakenly passing the `GetContext*` as the sixth argument and relying on the default (nullptr) for the seventh. This would make statistics inaccurate, at least.
Blame: 402b7aa0
Closes https://github.com/facebook/rocksdb/pull/3974
Differential Revision: D8344907
Pulled By: ajkr
fbshipit-source-id: 3ad865a0541d6d30f75dfc726352788118cfe12e
Summary:
Property block will be read sequentially and cached in a heap located
object, so there's no need for restart points. Thus we set the restart
interval to infinity to save space.
Closes https://github.com/facebook/rocksdb/pull/3970
Differential Revision: D8332586
Pulled By: fgwu
fbshipit-source-id: 899c3267832a81d0f084ec2db6b387332f461134
Summary:
BadOptions test creates a temporary db path changed to
table_block_based_bad_options_test to avoid collide with that created by
the PrefixAndWholeKeyTest
Closes https://github.com/facebook/rocksdb/pull/3965
Differential Revision: D8316080
Pulled By: fgwu
fbshipit-source-id: bb8e0fdfdb9abf0e5ce94494b4388cd1622ee032
Summary:
format_version 3 changes the format of index blocks by storing user keys instead of the internal keys, which saves 8-bytes per key. This patch extends the format to top-level indexes in partitioned index/filters.
Closes https://github.com/facebook/rocksdb/pull/3958
Differential Revision: D8294615
Pulled By: maysamyabandeh
fbshipit-source-id: 17666cc16b8076c363972e2308e31547e835f0fe
Summary:
PR https://github.com/facebook/rocksdb/pull/3838 made some changes that triggers lint warnings.
Run `make format` to fix formatting as suggested by siying .
Also piggyback two changes:
1) fix singleton destruction order for windows and posix env
2) fix two clang warnings
Closes https://github.com/facebook/rocksdb/pull/3954
Differential Revision: D8272041
Pulled By: miasantreble
fbshipit-source-id: 7c4fd12bd17aac13534520de0c733328aa3c6c9f
Summary:
This fixes a regression in one of myrocks regression tests (readwhilewriting), introduced in 8bf555f487
This PR changes two lines of code: one of them actually fixes the observed regression, the other is a mostly unrelated small fix that I'm piggy-backing here. EDIT: Nevermind, it fixes one line. More details in inline comments.
Closes https://github.com/facebook/rocksdb/pull/3953
Differential Revision: D8270664
Pulled By: al13n321
fbshipit-source-id: a7d91e196807d1e816551591257c700f70e4ccac
Summary:
format_version=3 changes the format of SST index. This is however not being tested currently since tests only work with the default format_version which is currently 2. The patch extends the most related tests to also test for format_version=3.
Closes https://github.com/facebook/rocksdb/pull/3942
Differential Revision: D8238413
Pulled By: maysamyabandeh
fbshipit-source-id: 915725f55753dd8e9188e802bf471c23645ad035
Summary:
Windows does not have LD_PRELOAD mechanism to override all memory allocation functions and ZSTD makes use of C-tuntime calloc. During flushes and compactions default system allocator fragments and the system slows down considerably.
For builds with jemalloc we employ an advanced ZSTD context creation API that re-directs memory allocation to jemalloc. To reduce the cost of context creation on each block we cache ZSTD context within the block based table builder while a new SST file is being built, this will help all platform builds including those w/o jemalloc. This avoids system allocator fragmentation and improves the performance.
The change does not address random reads and currently on Windows reads with ZSTD regress as compared with SNAPPY compression.
Closes https://github.com/facebook/rocksdb/pull/3838
Differential Revision: D8229794
Pulled By: miasantreble
fbshipit-source-id: 719b622ab7bf4109819bc44f45ec66f0dd3ee80d
Summary:
For iterator reads, a `SuperVersion` is pinned to preserve a snapshot of SST files, and `Block`s are pinned to allow `key()` and `value()` to return pointers directly into a RocksDB memory region. This works for both non-mmap reads, where the block owns the memory region, and mmap reads, where the file owns the memory region.
For point reads with `PinnableSlice`, only the `Block` object is pinned. This works for non-mmap reads because the block owns the memory region, so even if the file is deleted after compaction, the memory region survives. However, for mmap reads, file deletion causes the memory region to which the `PinnableSlice` refers to be unmapped. The result is usually a segfault upon accessing the `PinnableSlice`, although sometimes it returned wrong results (I repro'd this a bunch of times with `db_stress`).
This PR copies the value into the `PinnableSlice` when it comes from mmap'd memory. We can tell whether the `Block` owns its memory using `Block::cachable()`, which is unset when reads do not use the provided buffer as is the case with mmap file reads. When that is false we ensure the result of `Get()` is copied.
This feels like a short-term solution as ideally we'd have the `PinnableSlice` pin the mmap'd memory so we can do zero-copy reads. It seemed hard so I chose this approach to fix correctness in the meantime.
Closes https://github.com/facebook/rocksdb/pull/3881
Differential Revision: D8076288
Pulled By: ajkr
fbshipit-source-id: 31d78ec010198723522323dbc6ea325122a46b08
Summary:
This should resolve the performance regression caused by the unnecessary copying of the shared_ptr.
Closes https://github.com/facebook/rocksdb/pull/3937
Differential Revision: D8232330
Pulled By: miasantreble
fbshipit-source-id: 7885bf7cd190b6f87164c52d6edd328298c13f97
Summary:
DBTestBase::OptionConfig includes the scenarios that unit tests could iterate over them by calling ChangeOptions(). Some of the options have been mistakenly put after kEnd which makes them essentially invisible to ChangeOptions() caller. This patch fixes it except for kUniversalSubcompactions which is left as TODO since it would break some unit tests.
Closes https://github.com/facebook/rocksdb/pull/3935
Differential Revision: D8230748
Pulled By: maysamyabandeh
fbshipit-source-id: edddb8fffcd161af1809fef24798ce118f8593db
Summary:
The very old sst formats do not have table_properties and rep_->table_properties is thus nullptr. The recent patch in https://github.com/facebook/rocksdb/pull/3894 does not check for nullptr and hence makes it backward incompatible. This patch adds the check.
Closes https://github.com/facebook/rocksdb/pull/3918
Differential Revision: D8188638
Pulled By: maysamyabandeh
fbshipit-source-id: b1d986665ecf0b4d1c442adfa8a193b97707d47b
Summary:
Index blocks have the same format as data blocks. The keys therefore similarly to the keys in the data blocks are internal keys, which means that in addition to the user key it also has 8 bytes that encodes sequence number and value type. This extra 8 bytes however is not necessary in index blocks since the index keys act as an separator between two data blocks. The only exception is when the last key of a block and the first key of the next block share the same user key, in which the sequence number is required to act as a separator.
The patch excludes the sequence from index keys only if the above special case does not happen for any of the index keys. It then records that in the property block. The reader looks at the property block to see if it should expect sequence numbers in the keys of the index block.s
Closes https://github.com/facebook/rocksdb/pull/3894
Differential Revision: D8118775
Pulled By: maysamyabandeh
fbshipit-source-id: 915479f028b5799ca91671d67455ecdefbd873bd
Summary:
This was missed in a refactor of `ReadBlockContents` (2f1a3a4).
Closes https://github.com/facebook/rocksdb/pull/3906
Differential Revision: D8172648
Pulled By: ajkr
fbshipit-source-id: 27e453b19795fea974bfed4721105be6f3a12090
Summary:
Please refer to earlier discussion in [issue 3609](https://github.com/facebook/rocksdb/issues/3609).
There was also an alternative fix in [PR 3888](https://github.com/facebook/rocksdb/pull/3888), but the proposed solution requires complex change.
To summarize the cause of the problem. Upon creation of a column family, a `BlockBasedTableFactory` object is `new`ed and encapsulated by a `std::shared_ptr`. Since there is no other `std::shared_ptr` pointing to this `BlockBasedTableFactory`, when the column family is dropped, the `ColumnFamilyData` is `delete`d, causing the destructor of `std::shared_ptr`. Since there is no other `std::shared_ptr`, the underlying memory is also freed.
Later when the db exits, it releases all the table readers, including the table readers that have been operating on the dropped column family. This needs to access the `table_options` owned by `BlockBasedTableFactory` that has already been deleted. Therefore, a segfault is raised.
Previous workaround is to purge all obsolete files upon `ColumnFamilyData` destruction, which leads to a force release of table readers of the dropped column family. However this does not work when the user disables file deletion.
Our solution in this PR is making a copy of `table_options` in `BlockBasedTable::Rep`. This solution increases memory copy and usage, but is much simpler.
Test plan
```
$ make -j16
$ ./column_family_test --gtest_filter=ColumnFamilyTest.CreateDropAndDestroy:ColumnFamilyTest.CreateDropAndDestroyWithoutFileDeletion
```
Expected behavior:
All tests should pass.
Closes https://github.com/facebook/rocksdb/pull/3898
Differential Revision: D8149421
Pulled By: riversand963
fbshipit-source-id: eaecc2e064057ef607fbdd4cc275874f866c3438
Summary:
Currently when ldb built from master tries to open a DB from version 2.2, there will be a segfault because table_properties didn't exist back then.
Closes https://github.com/facebook/rocksdb/pull/3890
Differential Revision: D8100914
Pulled By: miasantreble
fbshipit-source-id: b255e8aedc54695432be2e704839c857dabdd65a
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:
`RangeDelAggregator` holds the pointers returned by `BlockIter::key()` and `BlockIter::value()` so requires the data to which they point is pinned. `BlockIter::key()` points into block memory and is guaranteed to be pinned if and only if prefix encoding is disabled (or, equivalently, restart interval is set to one). I think `BlockIter::value()` is always pinned. Added an assert for these and removed the wrong TODO about increasing restart interval, which would enable key prefix encoding and break the assertion.
Closes https://github.com/facebook/rocksdb/pull/3875
Differential Revision: D8063667
Pulled By: ajkr
fbshipit-source-id: 60b5ebcc0cdd610dd6aad9e74a23378793672c41
Summary:
Right now, every Block::NewIterator() reads num_restarts_ from the block, which is already read in Block::Block(). This sometimes cause a CPU cache miss. Although fetching this cacheline can usually benefit follow-up block restart offset reading, as they are close to each other, it's almost free to get ride of this read by storing it in the Block class.
Closes https://github.com/facebook/rocksdb/pull/3869
Differential Revision: D8052493
Pulled By: siying
fbshipit-source-id: 9c72360f0c2d7329f3c198ce4eaedd2bc14b87c1
Summary:
Before this PR, Iterator/InternalIterator may simultaneously have non-ok status() and Valid() = true. That state means that the last operation failed, but the iterator is nevertheless positioned on some unspecified record. Likely intended uses of that are:
* If some sst files are corrupted, a normal iterator can be used to read the data from files that are not corrupted.
* When using read_tier = kBlockCacheTier, read the data that's in block cache, skipping over the data that is not.
However, this behavior wasn't documented well (and until recently the wiki on github had misleading incorrect information). In the code there's a lot of confusion about the relationship between status() and Valid(), and about whether Seek()/SeekToLast()/etc reset the status or not. There were a number of bugs caused by this confusion, both inside rocksdb and in the code that uses rocksdb (including ours).
This PR changes the convention to:
* If status() is not ok, Valid() always returns false.
* Any seek operation resets status. (Before the PR, it depended on iterator type and on particular error.)
This does sacrifice the two use cases listed above, but siying said it's ok.
Overview of the changes:
* A commit that adds missing status checks in MergingIterator. This fixes a bug that actually affects us, and we need it fixed. `DBIteratorTest.NonBlockingIterationBugRepro` explains the scenario.
* Changes to lots of iterator types to make all of them conform to the new convention. Some bug fixes along the way. By far the biggest changes are in DBIter, which is a big messy piece of code; I tried to make it less big and messy but mostly failed.
* A stress-test for DBIter, to gain some confidence that I didn't break it. It does a few million random operations on the iterator, while occasionally modifying the underlying data (like ForwardIterator does) and occasionally returning non-ok status from internal iterator.
To find the iterator types that needed changes I searched for "public .*Iterator" in the code. Here's an overview of all 27 iterator types:
Iterators that didn't need changes:
* status() is always ok(), or Valid() is always false: MemTableIterator, ModelIter, TestIterator, KVIter (2 classes with this name anonymous namespaces), LoggingForwardVectorIterator, VectorIterator, MockTableIterator, EmptyIterator, EmptyInternalIterator.
* Thin wrappers that always pass through Valid() and status(): ArenaWrappedDBIter, TtlIterator, InternalIteratorFromIterator.
Iterators with changes (see inline comments for details):
* DBIter - an overhaul:
- It used to silently skip corrupted keys (`FindParseableKey()`), which seems dangerous. This PR makes it just stop immediately after encountering a corrupted key, just like it would for other kinds of corruption. Let me know if there was actually some deeper meaning in this behavior and I should put it back.
- It had a few code paths silently discarding subiterator's status. The stress test caught a few.
- The backwards iteration code path was expecting the internal iterator's set of keys to be immutable. It's probably always true in practice at the moment, since ForwardIterator doesn't support backwards iteration, but this PR fixes it anyway. See added DBIteratorTest.ReverseToForwardBug for an example.
- Some parts of backwards iteration code path even did things like `assert(iter_->Valid())` after a seek, which is never a safe assumption.
- It used to not reset status on seek for some types of errors.
- Some simplifications and better comments.
- Some things got more complicated from the added error handling. I'm open to ideas for how to make it nicer.
* MergingIterator - check status after every operation on every subiterator, and in some places assert that valid subiterators have ok status.
* ForwardIterator - changed to the new convention, also slightly simplified.
* ForwardLevelIterator - fixed some bugs and simplified.
* LevelIterator - simplified.
* TwoLevelIterator - changed to the new convention. Also fixed a bug that would make SeekForPrev() sometimes silently ignore errors from first_level_iter_.
* BlockBasedTableIterator - minor changes.
* BlockIter - replaced `SetStatus()` with `Invalidate()` to make sure non-ok BlockIter is always invalid.
* PlainTableIterator - some seeks used to not reset status.
* CuckooTableIterator - tiny code cleanup.
* ManagedIterator - fixed some bugs.
* BaseDeltaIterator - changed to the new convention and fixed a bug.
* BlobDBIterator - seeks used to not reset status.
* KeyConvertingIterator - some small change.
Closes https://github.com/facebook/rocksdb/pull/3810
Differential Revision: D7888019
Pulled By: al13n321
fbshipit-source-id: 4aaf6d3421c545d16722a815b2fa2e7912bc851d
Summary:
Now BlockBasedTableIterator directly uses BlockIter. By making BlockIter final, we can prevent unintended virtual function overriding.
Closes https://github.com/facebook/rocksdb/pull/3828
Differential Revision: D7933816
Pulled By: siying
fbshipit-source-id: 026a08cb5c5b6d3d6f44743152b4251da4756f2c
Summary:
Currently HarnessTest.Randomized is already split but some of the splits are faster than the others. The reason is that each split takes a continuous range of the generated args and the test with later args takes longer to finish. The patch evenly split the args among splits in a round robin fashion.
Before:
```
[ OK ] HarnessTest.Randomized1n2 (2278 ms)
[ OK ] HarnessTest.Randomized3n4 (1095 ms)
[ OK ] HarnessTest.Randomized5 (658 ms)
[ OK ] HarnessTest.Randomized6 (1258 ms)
[ OK ] HarnessTest.Randomized7 (6476 ms)
[ OK ] HarnessTest.Randomized8 (8182 ms)
```
After
```
[ OK ] HarnessTest.Randomized1 (2649 ms)
[ OK ] HarnessTest.Randomized2 (2645 ms)
[ OK ] HarnessTest.Randomized3 (2577 ms)
[ OK ] HarnessTest.Randomized4 (2490 ms)
[ OK ] HarnessTest.Randomized5 (2553 ms)
[ OK ] HarnessTest.Randomized6 (2560 ms)
[ OK ] HarnessTest.Randomized7 (2501 ms)
[ OK ] HarnessTest.Randomized8 (2574 ms)
```
Closes https://github.com/facebook/rocksdb/pull/3808
Differential Revision: D7882663
Pulled By: maysamyabandeh
fbshipit-source-id: 09b749a9684b6d7d65466aa4b00c5334a49e833e
Summary:
When the dummy record insertion fails, there is no need to explicitly delete the block as it will be registered for cleanup regardless.
Closes https://github.com/facebook/rocksdb/pull/3688
Differential Revision: D7537741
Pulled By: miasantreble
fbshipit-source-id: fcd3a3d3d382ee8e2c7ced0a4980e683d93a16d6
Summary:
7a6353bd1c prevents empty filter blocks from being written for SST files containing range deletions only. However the assertion this PR removes is still a problem as we could be reading from a DB generated by a RocksDB build without the 7a6353bd1c patch. So remove the assertion. We already don't do this check when `cache_index_and_filter_blocks=false`, so it should be safe.
Closes https://github.com/facebook/rocksdb/pull/3773
Differential Revision: D7769964
Pulled By: ajkr
fbshipit-source-id: 7285762446f2cd2ccf16efd7a988a106fbb0d8d3
Summary:
bc0da4b512 optimized bloom filters by skipping duplicate entires when the whole key and prefixes are both added to the bloom. It however used empty string as the initial value of the last entry added to the bloom. This is incorrect since empty key/prefix are valid entires by themselves. This patch fixes that.
Closes https://github.com/facebook/rocksdb/pull/3776
Differential Revision: D7778803
Pulled By: maysamyabandeh
fbshipit-source-id: d5a065daebee17f9403cac51e9d5626aac87bfbc
Summary:
Currently we rely on FilterBitsBuilder to skip the duplicate keys. It does that by comparing that hash of the key to the hash of the last added entry. This logic breaks however when we have whole_key_filtering mixed with prefix blooms as their addition to FilterBitsBuilder will be interleaved. The patch fixes that by comparing the last whole key and last prefix with the whole key and prefix of the new key respectively and skip the call to FilterBitsBuilder if it is a duplicate.
Closes https://github.com/facebook/rocksdb/pull/3764
Differential Revision: D7744413
Pulled By: maysamyabandeh
fbshipit-source-id: 15df73bbbafdfd754d4e1f42ea07f47b03bc5eb8
Summary:
Currently the partitioned index iterator creates a new ReadOptions which ignores the fill_cache config set to ReadOptions passed by the user. The patch propagates fill_cache from the user's ReadOptions to that of partition index iterator.
Also it clarifies the contract of fill_cache that i) it does not apply to filters, ii) it still charges block cache for the size of the data block, it still pin the block if it is already in the block cache.
Closes https://github.com/facebook/rocksdb/pull/3739
Differential Revision: D7678308
Pulled By: maysamyabandeh
fbshipit-source-id: 53ed96424ae922e499e2d4e3580ddc3f0db893da
Summary:
this PR fixes a few failed contbuild:
1. ASAN memory leak in Block::NewIterator (table/block.cc:429). the proper destruction of first_level_iter_ and second_level_iter_ of two_level_iterator.cc is missing from the code after the refactoring in https://github.com/facebook/rocksdb/pull/3406
2. various unused param errors introduced by https://github.com/facebook/rocksdb/pull/3662
3. updated comment for `ForceReleaseCachedEntry` to emphasize the use of `force_erase` flag.
Closes https://github.com/facebook/rocksdb/pull/3718
Reviewed By: maysamyabandeh
Differential Revision: D7621192
Pulled By: miasantreble
fbshipit-source-id: 476c94264083a0730ded957c29de7807e4f5b146
Summary:
This PR comments out the rest of the unused arguments which allow us to turn on the -Wunused-parameter flag. This is the second part of a codemod relating to https://github.com/facebook/rocksdb/pull/3557.
Closes https://github.com/facebook/rocksdb/pull/3662
Differential Revision: D7426121
Pulled By: Dayvedde
fbshipit-source-id: 223994923b42bd4953eb016a0129e47560f7e352
Summary:
The existing unit test did not set the level so the check for pinned partitioned filter/index being properly released from the block cache was not properly exercised as they only take effect in level 0. As a result a memory leak in pinned partitioned filters was hidden. The patch fix the test as well as the bug.
Closes https://github.com/facebook/rocksdb/pull/3692
Differential Revision: D7559763
Pulled By: maysamyabandeh
fbshipit-source-id: 55eff274945838af983c764a7d71e8daff092e4a
Summary:
Adds two stats to allow us measuring the false positive rate of full filters:
- The total count of positives: rocksdb.bloom.filter.full.positive
- The total count of true positives: rocksdb.bloom.filter.full.true.positive
Not the term "full" in the stat name to indicate that they are meaningful in full filters. block-based filters are to be deprecated soon and supporting it is not worth the the additional cost of if-then-else branches.
Closes#3680
Tested by:
$ ./db_bench -benchmarks=fillrandom -db /dev/shm/rocksdb-tmpdb --num=1000000 -bloom_bits=10
$ ./db_bench -benchmarks="readwhilewriting" -db /dev/shm/rocksdb-tmpdb --statistics -bloom_bits=10 --duration=60 --num=2000000 --use_existing_db 2>&1 > /tmp/full.log
$ grep filter.full /tmp/full.log
rocksdb.bloom.filter.full.positive COUNT : 3628593
rocksdb.bloom.filter.full.true.positive COUNT : 3536026
which gives the false positive rate of 2.5%
Closes https://github.com/facebook/rocksdb/pull/3681
Differential Revision: D7517570
Pulled By: maysamyabandeh
fbshipit-source-id: 630ab1a473afdce404916d297035b6318de4c052
Summary:
Our valgrind continuous test found an interesting leak which got introduced in #3614. We were adding the prefix key before saving the previous prefix start offset, due to which previous prefix offset is always incorrect. Fixed it by saving the the previous sate before adding the key.
Closes https://github.com/facebook/rocksdb/pull/3660
Differential Revision: D7418698
Pulled By: sagar0
fbshipit-source-id: 9933685f943cf2547ed5c553f490035a2fa785cf
Summary:
Provide a block_align option in BlockBasedTableOptions to allow
alignment of SST file data blocks. This will avoid higher
IOPS/throughput load due to < 4KB data blocks spanning 2 4KB pages.
When this option is set to true, the block alignment is set to lower of
block size and 4KB.
Closes https://github.com/facebook/rocksdb/pull/3502
Differential Revision: D7400897
Pulled By: anand1976
fbshipit-source-id: 04cc3bd144e88e3431a4f97604e63ad7a0f06d44
Summary:
b555ed30a4 introduces a regression, which causes blocks always to be pinned in block based iterators. Fix it.
Closes https://github.com/facebook/rocksdb/pull/3582
Differential Revision: D7189534
Pulled By: siying
fbshipit-source-id: 117dc7a03d0a0e360424db02efb366e12da2be03
Summary:
In attempting to build a static lib for use in iOS, I ran in to lots of type errors between uint64_t and size_t. This PR contains the changes I made to get `TARGET_OS=IOS make static_lib` to succeed while also getting Xcode to build successfully with the resulting `librocksdb.a` library imported.
This also compiles for me on macOS and tests fine, but I'm really not sure if I made the correct decisions about where to `static_cast` and where to change types.
Also up for discussion: is iOS worth supporting? Getting the static lib is just part one, we aren't providing any bridging headers or wrappers like the ObjectiveRocks project, it won't be a great experience.
Closes https://github.com/facebook/rocksdb/pull/3503
Differential Revision: D7106457
Pulled By: gfosco
fbshipit-source-id: 82ac2073de7e1f09b91f6b4faea91d18bd311f8e
Summary:
- removed a few unneeded variables
- fused some variable declarations and their assignments
- fixed right-trimming code in string_util.cc to not underflow
- simplifed an assertion
- move non-nullptr check assertion before dereferencing of that pointer
- pass an std::string function parameter by const reference instead of by value (avoiding potential copy)
Closes https://github.com/facebook/rocksdb/pull/3507
Differential Revision: D7004679
Pulled By: sagar0
fbshipit-source-id: 52944952d9b56dfcac3bea3cd7878e315bb563c4
Summary:
Use a customzied BlockBasedTableIterator and LevelIterator to replace current implementations leveraging two-level-iterator. Hope the customized logic will make code easier to understand. As a side effect, BlockBasedTableIterator reduces the allocation for the data block iterator object, and avoid the virtual function call to it, because we can directly reference BlockIter, a final class. Similarly, LevelIterator reduces virtual function call to the dummy iterator iterating the file metadata. It also enabled further optimization.
The upper bound check is also moved from index block to data block. This implementation fits this iterator better. After the change, forwared iterator is slightly optimized to ensure we trim those iterators.
The two-level-iterator now is only used by partitioned index, so it is simplified.
Closes https://github.com/facebook/rocksdb/pull/3406
Differential Revision: D6809041
Pulled By: siying
fbshipit-source-id: 7da3b9b1d3c8e9d9405302c15920af1fcaf50ffa
Summary:
`ReadBlockFromFile` uses a stack buffer to hold small data blocks before passing them to the compression library, which outputs uncompressed data in a heap buffer. In the case of `kNoCompression` there is a `memcpy` to copy from stack buffer to heap buffer.
This PR optimizes `ReadBlockFromFile` to skip the stack buffer for files whose blocks are known to be uncompressed. We determine this using the SST file property, "compression_name", if it's available.
Closes https://github.com/facebook/rocksdb/pull/3472
Differential Revision: D6920848
Pulled By: ajkr
fbshipit-source-id: 5c753e804efc178b9229ae5dbe6a4adc32031f07
Summary:
RandomizedHarnessTest enumerates different combinations of test type, compression type, restart interval, etc. For some combinations it takes very long to finish, causing the test to time out in test infrastructure.
This PR split the test input into smaller trunks in the hope that they will fit in the timeout window. Another possibility is to reduce `num_entries` of course
Closes https://github.com/facebook/rocksdb/pull/3467
Differential Revision: D6910235
Pulled By: miasantreble
fbshipit-source-id: 717246ee5d21a8a48ad82d4d9c04f9051a66f07f
Summary:
Previously `ReadBlockFromFile` for data blocks was only measured when reading a block to populate block cache. This PR adds the corresponding measurements for users who disabled block cache.
Closes https://github.com/facebook/rocksdb/pull/3442
Differential Revision: D6848671
Pulled By: ajkr
fbshipit-source-id: bb4bbe1797fa2cc1d9a5bad44891af2b55384b41
Summary:
ReadOptions.fill_cache is set in compaction inputs and can be set by users in their queries too. It tells RocksDB not to put a data block used to block cache.
The memory used by the data block is, however, not trackable by users.
To make the system more manageable, we can cost the block to block cache while using it, and then release it after using.
Closes https://github.com/facebook/rocksdb/pull/3333
Differential Revision: D6670230
Pulled By: miasantreble
fbshipit-source-id: ab848d3ed286bd081a13ee1903de357b56cbc308
Summary: Grandfather in super old lint issues to make a clean slate for moving forward that allows us to have stronger enforcement on new issues.
Reviewed By: yiwu-arbug
Differential Revision: D6821806
fbshipit-source-id: 22797d31ec58e9eb0255d3b66fedfcfcb0dc127c
Summary:
This change improves the performance of iterators doing long range scans (e.g. big/full table scans in MyRocks) by using readahead and prefetching additional data on each disk IO. This prefetching is automatically enabled on noticing more than 2 IOs for the same table file during iteration. The readahead size starts with 8KB and is exponentially increased on each additional sequential IO, up to a max of 256 KB. This helps in cutting down the number of IOs needed to complete the range scan.
Constraints:
- The prefetched data is stored by the OS in page cache. So this currently works only for non direct-reads use-cases i.e applications which use page cache. (Direct-I/O support will be enabled in a later PR).
- This gets currently enabled only when ReadOptions.readahead_size = 0 (which is the default value).
Thanks to siying for the original idea and implementation.
**Benchmarks:**
Data fill:
```
TEST_TMPDIR=/data/users/$USER/benchmarks/iter ./db_bench -benchmarks=fillrandom -num=1000000000 -compression_type="none" -level_compaction_dynamic_level_bytes
```
Do a long range scan: Seekrandom with large number of nexts
```
TEST_TMPDIR=/data/users/$USER/benchmarks/iter ./db_bench -benchmarks=seekrandom -duration=60 -num=1000000000 -use_existing_db -seek_nexts=10000 -statistics -histogram
```
Page cache was cleared before each experiment with the command:
```
sudo sh -c "echo 3 > /proc/sys/vm/drop_caches"
```
```
Before:
seekrandom : 34020.945 micros/op 29 ops/sec; 32.5 MB/s (1636 of 1999 found)
With this change:
seekrandom : 8726.912 micros/op 114 ops/sec; 126.8 MB/s (5702 of 6999 found)
```
~3.9X performance improvement.
Also verified with strace and gdb that the readahead size is increasing as expected.
```
strace -e readahead -f -T -t -p <db_bench process pid>
```
Closes https://github.com/facebook/rocksdb/pull/3282
Differential Revision: D6586477
Pulled By: sagar0
fbshipit-source-id: 8a118a0ed4594fbb7f5b1cafb242d7a4033cb58c
Summary:
This is a pre-cleaning up before a major block based table iterator refactoring. BlockBasedTable::NewDataBlockIterator() will always return BlockIter. This simplifies the logic and code and enable further refactoring and optimization.
Closes https://github.com/facebook/rocksdb/pull/3398
Differential Revision: D6780165
Pulled By: siying
fbshipit-source-id: 273f7dc896724f682c0118fb69a359d9cc4418b4
Summary:
* Fix DBTest.CompactRangeWithEmptyBottomLevel lite build failure
* Fix DBTest.AutomaticConflictsWithManualCompaction failure introduce by #3366
* Fix BlockBasedTableTest::IndexUncompressed should be disabled if snappy is disabled
* Fix ASAN failure with DBBasicTest::DBClose test
Closes https://github.com/facebook/rocksdb/pull/3373
Differential Revision: D6732313
Pulled By: yiwu-arbug
fbshipit-source-id: 1eb9b9d9a8d795f56188fa9770db9353f6fdedc5
Summary:
Re-use metadata for reading Compression Dictionary on BlockBased
table open, this saves two reads from disk.
This helps to our 999 percentile in 5.6.1 where prefetch buffer is not present.
Closes https://github.com/facebook/rocksdb/pull/3354
Differential Revision: D6695753
Pulled By: ajkr
fbshipit-source-id: bb8acd9e9e66e65b89c548ab8940570ae360333c
Summary:
BlockTest.BlockReadAmpBitmap is too slow and times out in some environments. Speed it up by:
(1) improve the way the verification is done. With this it is 5 times faster
(2) run fewer tests for large blocks. This cut it down by another 10 times.
Now it can finish in similar time as other tests.
Closes https://github.com/facebook/rocksdb/pull/3313
Differential Revision: D6643711
Pulled By: siying
fbshipit-source-id: c2397d666eab5421a78ca87e1e45491e0f832a6d
Summary:
NUMBER_BLOCK_DECOMPRESSED and NUMBER_BLOCK_COMPRESSED are not reported unless the stats level contain detailed timers, which is wrong. They are normal counters. Fix it.
Closes https://github.com/facebook/rocksdb/pull/3263
Differential Revision: D6552519
Pulled By: siying
fbshipit-source-id: 40899ccea7b2856bb39752616657c0bfd432f6f9
Summary:
Some call sites of BlockFetcher create temporary ReadOptions and pass to BlockFetcher. The temporary object will be gone after BlockFetcher construction but BlockFetcher keep its reference, causing stack-use-after-scope. Fixing it.
Closes https://github.com/facebook/rocksdb/pull/3258
Differential Revision: D6547152
Pulled By: yiwu-arbug
fbshipit-source-id: 6b49e9dd46bb72307f5d8f88ea15faacff35b9bc
Summary:
Divide ReadBlockContents() to multiple sub-functions. Maintaining the input and intermediate data in a new class BlockFetcher.
I hope in general it makes the code easier to maintain.
Another motivation to do it is to clearly divide the logic before file reading and after file reading. The refactor will help us evaluate how can we make I/O async in the future.
Closes https://github.com/facebook/rocksdb/pull/3244
Differential Revision: D6520983
Pulled By: siying
fbshipit-source-id: 338d90bc0338472d46be7a7682028dc9114b12e9
Summary:
table/block.cc:
420 }
CID 1396127 (#1 of 1): Uninitialized scalar field (UNINIT_CTOR)
7. uninit_member: Non-static class member restart_offset_ is not initialized in this constructor nor in any functions that it calls.
421}
table/block_based_table_builder.cc:
CID 1418259 (#1 of 1): Uninitialized scalar field (UNINIT_CTOR)
7. uninit_member: Non-static class member compressed_cache_key_prefix_size is not initialized in this constructor nor in any functions that it calls.
table/block_based_table_reader.h:
3. uninit_member: Non-static class member index_type is not initialized in this constructor nor in any functions that it calls.
CID 1396147 (#1 of 1): Uninitialized scalar field (UNINIT_CTOR)
5. uninit_member: Non-static class member hash_index_allow_collision is not initialized in this constructor nor in any functions that it calls.
413 global_seqno(kDisableGlobalSequenceNumber) {}
414
table/cuckoo_table_reader.cc:
55 if (hash_funs == user_props.end()) {
56 status_ = Status::Corruption("Number of hash functions not found");
5. uninit_member: Non-static class member is_last_level_ is not initialized in this constructor nor in any functions that it calls.
7. uninit_member: Non-static class member identity_as_first_hash_ is not initialized in this constructor nor in any functions that it calls.
9. uninit_member: Non-static class member use_module_hash_ is not initialized in this constructor nor in any functions that it calls.
11. uninit_member: Non-static class member num_hash_func_ is not initialized in this constructor nor in any functions that it calls.
13. uninit_member: Non-static class member key_length_ is not initialized in this constructor nor in any functions that it calls.
15. uninit_member: Non-static class member user_key_length_ is not initialized in this constructor nor in any functions that it calls.
17. uninit_member: Non-static class member value_length_ is not initialized in this constructor nor in any functions that it calls.
19. uninit_member: Non-static class member bucket_length_ is not initialized in this constructor nor in any functions that it calls.
21. uninit_member: Non-static class member cuckoo_block_size_ is not initialized in this constructor nor in any functions that it calls.
23. uninit_member: Non-static class member cuckoo_block_bytes_minus_one_ is not initialized in this constructor nor in any functions that it calls.
CID 1322785 (#2 of 2): Uninitialized scalar field (UNINIT_CTOR)
25. uninit_member: Non-static class member table_size_ is not initialized in this constructor nor in any functions that it calls.
57 return;
table/plain_table_index.h:
2. uninit_member: Non-static class member index_size_ is not initialized in this constructor nor in any functions that it calls.
CID 1322801 (#1 of 1): Uninitialized scalar field (UNINIT_CTOR)
4. uninit_member: Non-static class member sub_index_size_ is not initialized in this constructor nor in any functions that it calls.
128 huge_page_tlb_size_(huge_page_tlb_size) {}
129
Closes https://github.com/facebook/rocksdb/pull/3113
Differential Revision: D6505719
Pulled By: yiwu-arbug
fbshipit-source-id: 38f44d8f9dfefb4c2e25d83b8df25a5201c75618
Summary:
I started adding gflags support for cmake on linux and got frustrated that I'd need to duplicate the build_detect_platform logic, which determines namespace based on attempting compilation. We can do it differently -- use the GFLAGS_NAMESPACE macro if available, and if not, that indicates it's an old gflags version without configurable namespace so we can simply hardcode "google".
Closes https://github.com/facebook/rocksdb/pull/3212
Differential Revision: D6456973
Pulled By: ajkr
fbshipit-source-id: 3e6d5bde3ca00d4496a120a7caf4687399f5d656
Summary:
The ASCII output is almost always useless to me as the first '\0' byte in the key or value causes it to stop printing. Since all characters are already surrounded by spaces, "\ 0" (how we display a backslash followed by a zero) and "\0" (how this PR displays a null terminator) are distinguishable. My assumption is the value of seeing all the bytes outweighs the value of the alignment we had before, where we always had one character followed by one space.
Closes https://github.com/facebook/rocksdb/pull/3203
Differential Revision: D6428651
Pulled By: ajkr
fbshipit-source-id: aafc978a51e9ea029cfe3e763e2bb0e1751b9ccf
Summary:
Problem: Option string accepts only cache_size as parameter for block_cache which is specified as "block_cache=1M".
It doesn't accept other parameters like num_shards etc.
Changes :
1) ParseBlockBasedTableOption in block_based_table_factory is edited to accept cache options in the format "block_cache=<cache_size>:<num_shard_bits>:<strict_capacity_limit>:<high_pri_pool_ratio>".
Options other than cache_size are optional to maintain backward compatibility. The changes are valid for block_cache_compressed as well.
For example, "block_cache=1M:6:true:0.5", "block_cache=1M:6:true", "block_cache=1M:6" and "block_cache=1M" are all valid option strings.
2) Corresponding unit tests are added.
Closes https://github.com/facebook/rocksdb/pull/3108
Differential Revision: D6420997
Pulled By: sagar0
fbshipit-source-id: cdea8b785688d2802907974af27225ccc1c0cd43
Summary:
block_size_deviation is in percentage while the partition size is in bytes. The current code fails to take that into account resulting into very large target size for filter partitions.
Closes https://github.com/facebook/rocksdb/pull/3187
Differential Revision: D6376069
Pulled By: maysamyabandeh
fbshipit-source-id: 276546fc68f50e0da32c462abb46f6cf676db9b2
Summary:
We don't propagate TableProperty::oldest_key_time on compaction and just write the default value to SST files. It is more natural to default the value to 0.
Also revert db_sst_test back to before #2842.
Closes https://github.com/facebook/rocksdb/pull/3079
Differential Revision: D6165702
Pulled By: yiwu-arbug
fbshipit-source-id: ca3ce5928d96ae79a5beb12bb7d8c640a71478a0
Summary:
With FIFO compaction we would like to get the oldest data time for monitoring. The problem is we don't have timestamp for each key in the DB. As an approximation, we expose the earliest of sst file "creation_time" property.
My plan is to override the property with a more accurate value with blob db, where we actually have timestamp.
Closes https://github.com/facebook/rocksdb/pull/2842
Differential Revision: D5770600
Pulled By: yiwu-arbug
fbshipit-source-id: 03833c8f10bbfbee62f8ea5c0d03c0cafb5d853a
Summary:
Currently, RocksDB does not allow reopening a preexisting DB with no merge operator defined, with a merge operator defined. This means that if a DB ever want to add a merge operator, there's no way to do so currently.
Fix this by adding a new verification type `kByNameAllowFromNull` which will allow old values to be nullptr, and new values to be non-nullptr.
Closes https://github.com/facebook/rocksdb/pull/2958
Differential Revision: D5961131
Pulled By: lth
fbshipit-source-id: 06179bebd0d90db3d43690b5eb7345e2d5bab1eb
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:
In SST files, restart interval helps us search in data blocks. However, some meta blocks will be read sequentially, so there's no need for restart points. Restart interval will introduce extra space in the block (https://github.com/facebook/rocksdb/blob/master/table/block_builder.cc#L80). We will see if we can remove this redundant space. (Maybe set restart interval to infinite.)
Closes https://github.com/facebook/rocksdb/pull/2940
Differential Revision: D5930139
Pulled By: miasantreble
fbshipit-source-id: 92b1b23c15cffa90378343ac846b713623b19c21
Summary:
When using with compressed cache it is possible that the status is ok but the block is not actually added to the block cache. The patch takes this case into account.
Closes https://github.com/facebook/rocksdb/pull/2945
Differential Revision: D5937613
Pulled By: maysamyabandeh
fbshipit-source-id: 5428cf1115e5046b3d01ab78d26cb181122af4c6
Summary:
For every merge operand encountered for a key in the read path we now have the ability to decide whether to look further (to retrieve more merge operands for the key) or stop and invoke the merge operator to return the value. The user needs to override `ShouldMerge()` method with a condition to terminate search when true to avail this facility.
This has a couple of advantages:
1. It helps in limiting the number of merge operands that are looked at to compute a value as part of a user Get operation.
2. It allows to peek at a merge key-value to see if further merge operands need to look at.
Example: Limiting the number of merge operands that are looked at: Lets say you have 10 merge operands for a key spread over various levels. If you only want RocksDB to look at the latest two merge operands instead of all 10 to compute the value, it is now possible with this PR. You can set the condition in `ShouldMerge()` to return true when the size of the operand list is 2. Look at the example implementation in the unit test. Without this PR, a Get might look at all the 10 merge operands in different levels before invoking the merge-operator.
Added a new unit test.
Made sure that there is no perf regression by running benchmarks.
Command line to Load data:
```
TEST_TMPDIR=/dev/shm ./db_bench --benchmarks="mergerandom" --merge_operator="uint64add" --num=10000000
...
mergerandom : 12.861 micros/op 77757 ops/sec; 8.6 MB/s ( updates:10000000)
```
**ReadRandomMergeRandom bechmark results:**
Command line:
```
TEST_TMPDIR=/dev/shm ./db_bench --benchmarks="readrandommergerandom" --merge_operator="uint64add" --num=10000000
```
Base -- Without this code change (on commit fc7476b):
```
readrandommergerandom : 38.586 micros/op 25916 ops/sec; (reads:3001599 merges:6998401 total:10000000 hits:842235 maxlength:8)
```
With this code change:
```
readrandommergerandom : 38.653 micros/op 25870 ops/sec; (reads:3001599 merges:6998401 total:10000000 hits:842235 maxlength:8)
```
Closes https://github.com/facebook/rocksdb/pull/2923
Differential Revision: D5898239
Pulled By: sagar0
fbshipit-source-id: daefa325019f77968639a75c851d46352c2303ef
Summary:
A user encountered segfault on the call to `CacheDependencies()`, probably because `NewIndexIterator()` failed before populating `*index_entry`. Let's avoid the call in that case.
Closes https://github.com/facebook/rocksdb/pull/2939
Differential Revision: D5928611
Pulled By: ajkr
fbshipit-source-id: 484be453dbb00e5e160e9c6a1bc933df7d80f574
Summary:
Three small optimizations:
(1) iter_->IsKeyPinned() shouldn't be called if read_options.pin_data is not true. This may trigger function call all the way down the iterator tree.
(2) reuse the iterator key object in DBIter::FindNextUserEntryInternal(). The constructor of the class has some overheads.
(3) Move the switching direction logic in MergingIterator::Next() to a separate function.
These three in total improves readseq performance by about 3% in my benchmark setting.
Closes https://github.com/facebook/rocksdb/pull/2880
Differential Revision: D5829252
Pulled By: siying
fbshipit-source-id: 991aea10c6d6c3b43769cb4db168db62954ad1e3
Summary:
Merging iterator invokes InternalKeyComparator.Compare() frequently to heap merge. By making InternalKeyComparator final and merging iterator to directly use InternalKeyComparator rather than through Iterator interface, we can give compiler a choice to avoid one more virtual function call if possible. I ran readseq benchmark in memory-only use case to make sure the performance at least doesn't regress.
I have to disable the final key word in debug build, as a hack test class depends on overriding the class.
Closes https://github.com/facebook/rocksdb/pull/2860
Differential Revision: D5800461
Pulled By: siying
fbshipit-source-id: ab876f22a09bb5c560740911412336e0e25ccb53
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:
store a zero as the checksum when disabled since it's easier to keep block trailer a fixed length.
Closes https://github.com/facebook/rocksdb/pull/2781
Differential Revision: D5694702
Pulled By: ajkr
fbshipit-source-id: 69cea9da415778ba2b600dfd9d0dfc8cb5188ecd
Summary:
This is the warning that clang considers a bug and has been causing it to fail:
```
table/block_based_table_reader.cc:240:27: warning: Potential leak of memory pointed to by 'block.value'
for (; biter.Valid(); biter.Next()) {
^~~~~
```
Actually clang just doesn't have enough knowledge to statically determine it's safe. We can teach it using an assert.
Closes https://github.com/facebook/rocksdb/pull/2779
Differential Revision: D5691225
Pulled By: ajkr
fbshipit-source-id: 3f0d545bf44636953b30ee5243c63239e8f16d8e
Summary:
Allow `Slice` holding nullptr as a sentinel value but not in comparisons. This new restriction eliminates the need for the manual checks in 39ef900551, while still conforming to glibc's `memcmp` API. Thanks siying for the idea. Users may need to migrate, so mentioned it in HISTORY.md.
Closes https://github.com/facebook/rocksdb/pull/2777
Differential Revision: D5686016
Pulled By: ajkr
fbshipit-source-id: 03a2ca3fd9a0ebade9d0d5686c81d59a9534f563
Summary:
This is the continuation of https://github.com/facebook/rocksdb/pull/2661 for filter partitions. When pin_l0 is set (along with cache_xxx), then open table open the filter partitions are loaded into the cache and pinned there.
Closes https://github.com/facebook/rocksdb/pull/2766
Differential Revision: D5671098
Pulled By: maysamyabandeh
fbshipit-source-id: 174f24018f1d7f1129621e7380287b65b67d2115
Summary:
This fixes the existing logic for pinning l0 index partitions. The patch preloads the partitions into block cache and pin them if they belong to level 0 and pin_l0 is set.
The drawback is that it does many small IOs when preloading all the partitions into the cache is direct io is enabled. Working for a solution for that.
Closes https://github.com/facebook/rocksdb/pull/2661
Differential Revision: D5554010
Pulled By: maysamyabandeh
fbshipit-source-id: 1e6f32a3524d71355c77d4138516dcfb601ca7b2
Summary:
Right now, if direct I/O is enabled, prefetching the last 512KB cannot be applied, except compaction inputs or readahead is enabled for iterators. This can create a lot of I/O for HDD cases. To solve the problem, the 512KB is prefetched in block based table if direct I/O is enabled. The prefetched buffer is passed in totegher with random access file reader, so that we try to read from the buffer before reading from the file. This can be extended in the future to support flexible user iterator readahead too.
Closes https://github.com/facebook/rocksdb/pull/2708
Differential Revision: D5593091
Pulled By: siying
fbshipit-source-id: ee36ff6d8af11c312a2622272b21957a7b5c81e7
Summary:
We need a tool to check any sst file corruption in the db.
It will check all the sst files in current version and read all the blocks (data, meta, index) with checksum verification. If any verification fails, the function will return non-OK status.
Closes https://github.com/facebook/rocksdb/pull/2498
Differential Revision: D5324269
Pulled By: lightmark
fbshipit-source-id: 6f8a272008b722402a772acfc804524c9d1a483b
Summary:
Replace dynamic_cast<> so that users can choose to build with RTTI off, so that they can save several bytes per object, and get tiny more memory available.
Some nontrivial changes:
1. Add Comparator::GetRootComparator() to get around the internal comparator hack
2. Add the two experiemental functions to DB
3. Add TableFactory::GetOptionString() to avoid unnecessary casting to get the option string
4. Since 3 is done, move the parsing option functions for table factory to table factory files too, to be symmetric.
Closes https://github.com/facebook/rocksdb/pull/2645
Differential Revision: D5502723
Pulled By: siying
fbshipit-source-id: fd13cec5601cf68a554d87bfcf056f2ffa5fbf7c
Summary:
Breaking commit: d12691b86f
In the above commit, I moved the `TableCache` cleanup logic from `Version` destructor into `PurgeObsoleteFiles`. I missed cleaning up `TableCache` entries for the current `Version` during DB destruction.
This PR adds that logic to `VersionSet` destructor. One unfortunate side effect is now we're potentially deleting `TableReader`s after `column_family_set_.reset()`, which means we can't call `BlockBasedTableReader::Close` a second time as the block cache might already be destroyed.
Closes https://github.com/facebook/rocksdb/pull/2662
Differential Revision: D5515108
Pulled By: ajkr
fbshipit-source-id: 2cb820e19aa813e0d258d17f76b2d7b6b7ee0b18
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:
This patch enables using PinnableSlice for RowCache, changes include
not releasing the cache handle immediately after lookup in TableCache::Get, instead pass a Cleanble function which does Cache::RleaseHandle.
Closes https://github.com/facebook/rocksdb/pull/2492
Differential Revision: D5316216
Pulled By: maysamyabandeh
fbshipit-source-id: d2a684bd7e4ba73772f762e58a82b5f4fbd5d362
Summary:
In gcc-7 the following is an error identified by -Werror=class-memaccess
In file included from ./table/get_context.h:14:0,
from db/version_set.cc:43:
./table/block.h: In constructor ‘rocksdb::BlockReadAmpBitmap::BlockReadAmpBitmap(size_t, size_t, rocksdb::Statistics*)’:
./table/block.h:73:53: error: ‘void* memset(void*, int, size_t)’ clearing an object of type ‘struct std::atomic<unsigned int>’ with no trivial copy-assignment; use value-initialization instead [-Werror=class-memaccess]
memset(bitmap_, 0, bitmap_size * kBytesPersEntry);
^
In file included from ./db/version_set.h:23:0,
from db/version_set.cc:12:
/toolchain/include/c++/8.0.0/atomic:684:12: note: ‘struct std::atomic<unsigned int>’ declared here
struct atomic<unsigned int> : __atomic_base<unsigned int>
^~~~~~~~~~~~~~~~~~~~
As a solution the default initializer can be applied in list context.
Signed-off-by: Daniel Black <daniel.black@au.ibm.com>
Closes https://github.com/facebook/rocksdb/pull/2561
Differential Revision: D5398714
Pulled By: siying
fbshipit-source-id: d883fb88ec7535eee60d551038fe91f14488be36
Summary:
this modify allows third-party tables able to support delete range
Closes https://github.com/facebook/rocksdb/pull/2035
Differential Revision: D5407973
Pulled By: ajkr
fbshipit-source-id: 82e364b7dd5a198660788d59543f15b8f95cc418
Summary:
The casting seemed to cause a problem.
I think this might increase it to unsigned long.
Closes https://github.com/facebook/rocksdb/pull/2562
Differential Revision: D5406842
Pulled By: siying
fbshipit-source-id: 736adef31448229a58a1a48bdbe77792f36736e8
Summary:
Valgrind reports that it is not initialized.
Closes https://github.com/facebook/rocksdb/pull/2541
Differential Revision: D5376084
Pulled By: maysamyabandeh
fbshipit-source-id: 55c312f4f506863aa0d25ff92c8c34b57f48b860
Summary:
Currently metadata_block_size controls only index partition size. With this patch a partition is cut after any of index or filter partitions reaches metadata_block_size.
Closes https://github.com/facebook/rocksdb/pull/2452
Differential Revision: D5275651
Pulled By: maysamyabandeh
fbshipit-source-id: 5057e4424b4c8902043782e6bf8c38f0c4f25160
Summary:
We've got some DBs where iterators return Status with message "Corruption: block checksum mismatch" all the time. That's not very informative. It would be much easier to investigate if the error message contained the file name - then we would know e.g. how old the corrupted file is, which would be very useful for finding the root cause. This PR adds file name, offset and other stuff to some block corruption-related status messages.
It doesn't improve all the error messages, just a few that were easy to improve. I'm mostly interested in "block checksum mismatch" and "Bad table magic number" since they're the only corruption errors that I've ever seen in the wild.
Closes https://github.com/facebook/rocksdb/pull/2507
Differential Revision: D5345702
Pulled By: al13n321
fbshipit-source-id: fc8023d43f1935ad927cef1b9c55481ab3cb1339
Summary:
Introducing FIFO compactions with TTL.
FIFO compaction is based on size only which makes it tricky to enable in production as use cases can have organic growth. A user requested an option to drop files based on the time of their creation instead of the total size.
To address that request:
- Added a new TTL option to FIFO compaction options.
- Updated FIFO compaction score to take TTL into consideration.
- Added a new table property, creation_time, to keep track of when the SST file is created.
- Creation_time is set as below:
- On Flush: Set to the time of flush.
- On Compaction: Set to the max creation_time of all the files involved in the compaction.
- On Repair and Recovery: Set to the time of repair/recovery.
- Old files created prior to this code change will have a creation_time of 0.
- FIFO compaction with TTL is enabled when ttl > 0. All files older than ttl will be deleted during compaction. i.e. `if (file.creation_time < (current_time - ttl)) then delete(file)`. This will enable cases where you might want to delete all files older than, say, 1 day.
- FIFO compaction will fall back to the prior way of deleting files based on size if:
- the creation_time of all files involved in compaction is 0.
- the total size (of all SST files combined) does not drop below `compaction_options_fifo.max_table_files_size` even if the files older than ttl are deleted.
This feature is not supported if max_open_files != -1 or with table formats other than Block-based.
**Test Plan:**
Added tests.
**Benchmark results:**
Base: FIFO with max size: 100MB ::
```
svemuri@dev15905 ~/rocksdb (fifo-compaction) $ TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=readwhilewriting --num=5000000 --threads=16 --compaction_style=2 --fifo_compaction_max_table_files_size_mb=100
readwhilewriting : 1.924 micros/op 519858 ops/sec; 13.6 MB/s (1176277 of 5000000 found)
```
With TTL (a low one for testing) ::
```
svemuri@dev15905 ~/rocksdb (fifo-compaction) $ TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=readwhilewriting --num=5000000 --threads=16 --compaction_style=2 --fifo_compaction_max_table_files_size_mb=100 --fifo_compaction_ttl=20
readwhilewriting : 1.902 micros/op 525817 ops/sec; 13.7 MB/s (1185057 of 5000000 found)
```
Example Log lines:
```
2017/06/26-15:17:24.609249 7fd5a45ff700 (Original Log Time 2017/06/26-15:17:24.609177) [db/compaction_picker.cc:1471] [default] FIFO compaction: picking file 40 with creation time 1498515423 for deletion
2017/06/26-15:17:24.609255 7fd5a45ff700 (Original Log Time 2017/06/26-15:17:24.609234) [db/db_impl_compaction_flush.cc:1541] [default] Deleted 1 files
...
2017/06/26-15:17:25.553185 7fd5a61a5800 [DEBUG] [db/db_impl_files.cc:309] [JOB 0] Delete /dev/shm/dbbench/000040.sst type=2 #40 -- OK
2017/06/26-15:17:25.553205 7fd5a61a5800 EVENT_LOG_v1 {"time_micros": 1498515445553199, "job": 0, "event": "table_file_deletion", "file_number": 40}
```
SST Files remaining in the dbbench dir, after db_bench execution completed:
```
svemuri@dev15905 ~/rocksdb (fifo-compaction) $ ls -l /dev/shm//dbbench/*.sst
-rw-r--r--. 1 svemuri users 30749887 Jun 26 15:17 /dev/shm//dbbench/000042.sst
-rw-r--r--. 1 svemuri users 30768779 Jun 26 15:17 /dev/shm//dbbench/000044.sst
-rw-r--r--. 1 svemuri users 30757481 Jun 26 15:17 /dev/shm//dbbench/000046.sst
```
Closes https://github.com/facebook/rocksdb/pull/2480
Differential Revision: D5305116
Pulled By: sagar0
fbshipit-source-id: 3e5cfcf5dd07ed2211b5b37492eb235b45139174
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 currently do not support partitioning filters if indexes are not partitioned. The patch makes sure that these two are consistent.
Closes https://github.com/facebook/rocksdb/pull/2455
Differential Revision: D5275644
Pulled By: maysamyabandeh
fbshipit-source-id: b61701ac8914c2206d06f5e33ff6f67b24406d1d
Summary:
When Partitioning index/filter is enabled the user might need to check the index block size as well as the top-level index size via sst_dump. This patch records i) number of partitions, ii) top-level index size and make it accessible through sst_dump. The number of partitions for filters is the same as that of indexes. The top-level index for filters has a similar size to top-level index for indexes, so it is not repeated.
Closes https://github.com/facebook/rocksdb/pull/2437
Differential Revision: D5224225
Pulled By: maysamyabandeh
fbshipit-source-id: 5324598c75793523aef1bb7ee225a5475e95a9cb
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:
filter_block_set_ access must also be protected with mutex.
Closes https://github.com/facebook/rocksdb/pull/2413
Differential Revision: D5193159
Pulled By: maysamyabandeh
fbshipit-source-id: 6987fc219d9a65c20b9c7e52151aef4b8e4882e6
Summary:
… headers
https://github.com/facebook/rocksdb/pull/2199 should not reference RocksDB-specific macros (like ROCKSDB_SUPPORT_THREAD_LOCAL in this case) to public headers, `iostats_context.h` and `perf_context.h`. We shouldn't do that because users have to provide these compiler flags when building their binary with RocksDB.
We should hide the thread local global variable inside our implementation and just expose a function api to retrieve these variables. It may break some users for now but good for long term.
make check -j64
Closes https://github.com/facebook/rocksdb/pull/2380
Differential Revision: D5177896
Pulled By: lightmark
fbshipit-source-id: 6fcdfac57f2e2dcfe60992b7385c5403f6dcb390
Summary:
Fixes the following scenario:
1. Set prefix extractor. Enable bloom filters, with `whole_key_filtering = false`. Use compaction filter that sometimes returns `kRemoveAndSkipUntil`.
2. Do a compaction.
3. Compaction creates an iterator with `total_order_seek = false`, calls `SeekToFirst()` on it, then repeatedly calls `Next()`.
4. At some point compaction filter returns `kRemoveAndSkipUntil`.
5. Compaction calls `Seek(skip_until)` on the iterator. The key that it seeks to happens to have prefix that doesn't match the bloom filter. Since `total_order_seek = false`, iterator becomes invalid, and compaction thinks that it has reached the end. The rest of the compaction input is silently discarded.
The fix is to make compaction iterator use `total_order_seek = true`.
The implementation for PlainTable is quite awkward. I've made `kRemoveAndSkipUntil` officially incompatible with PlainTable. If you try to use them together, compaction will fail, and DB will enter read-only mode (`bg_error_`). That's not a very graceful way to communicate a misconfiguration, but the alternatives don't seem worth the implementation time and complexity. To be able to check in advance that `kRemoveAndSkipUntil` is not going to be used with PlainTable, we'd need to extend the interface of either `CompactionFilter` or `InternalIterator`. It seems unlikely that anyone will ever want to use `kRemoveAndSkipUntil` with PlainTable: PlainTable probably has very few users, and `kRemoveAndSkipUntil` has only one user so far: us (logdevice).
Closes https://github.com/facebook/rocksdb/pull/2349
Differential Revision: D5110388
Pulled By: lightmark
fbshipit-source-id: ec29101a99d9dcd97db33923b87f72bce56cc17a
Summary:
Some users want to monitor column family activity in their custom memtable implementations. Previously there was no way to figure out with which column family a memtable is associated. This diff:
- adds an overload to MemTableRepFactory::CreateMemTableRep() that provides the CF ID. For compatibility, its default implementation calls the old overload.
- updates MemTable to create MemTableRep's using the new overload.
Closes https://github.com/facebook/rocksdb/pull/2346
Differential Revision: D5108061
Pulled By: ajkr
fbshipit-source-id: 3a1921214a348dd8ea0f54e1cab3b71c3d46d616
Summary:
Previously sst_file_writer only supports kTypeValue, we need kTypeMerge and kTypeDeletion also as user requested.
Closes https://github.com/facebook/rocksdb/pull/2361
Differential Revision: D5139402
Pulled By: lightmark
fbshipit-source-id: 092a60756d01692539d817a3765ebfd58a8d7f88
Summary:
The default IO priority of WritableFiles is IO_TOTAL, meaning that
they will bypass the rate limiter if it's passed in the options.
This change allows to pass an io priority in construction, so that by
setting IO_LOW or IO_HIGH the rate limit will be honored.
It also fixes a minor bug: SstFileWriter's copy and move constructor
are not disabled and incorrect, as any copy/move will result in a
double free. Switching to unique_ptr makes the object correctly
movable and non-copyable as expected.
Also fix minor style inconsistencies.
Closes https://github.com/facebook/rocksdb/pull/2335
Differential Revision: D5113260
Pulled By: sagar0
fbshipit-source-id: e084236e7ff0b50a56cbeceaa9fedd5e210bf9f8
Summary:
BlockBasedTable::compaction_optimized_ is never used but can cause TSAN warning. Remove it.
Closes https://github.com/facebook/rocksdb/pull/2324
Differential Revision: D5085533
Pulled By: siying
fbshipit-source-id: 2feefce6806d559dfb4ab2989aa3db36752fe25d
Summary:
Consider BlockReadAmpBitmap with bytes_per_bit = 32. Suppose bytes [a, b) were used, while bytes [a-32, a)
and [b+1, b+33) weren't used; more formally, the union of ranges passed to BlockReadAmpBitmap::Mark() contains [a, b) and doesn't intersect with [a-32, a) and [b+1, b+33). Then bits [floor(a/32), ceil(b/32)] will be set, and so the number of useful bytes will be estimated as (ceil(b/32) - floor(a/32)) * 32, which is on average equal to b-a+31.
An extreme example: if we use 1 byte from each block, it'll be counted as 32 bytes from each block.
It's easy to remove this bias by slightly changing the semantics of the bitmap. Currently each bit represents a byte range [i*32, (i+1)*32).
This diff makes each bit represent a single byte: i*32 + X, where X is a random number in [0, 31] generated when bitmap is created. So, e.g., if you read a single byte at random, with probability 31/32 it won't be counted at all, and with probability 1/32 it will be counted as 32 bytes; so, on average it's counted as 1 byte.
*But there is one exception: the last bit will always set with the old way.*
(*) - assuming read_amp_bytes_per_bit = 32.
Closes https://github.com/facebook/rocksdb/pull/2259
Differential Revision: D5035652
Pulled By: lightmark
fbshipit-source-id: bd98b1b9b49fbe61f9e3781d07f624e3cbd92356
Summary:
Based on my experience with linkbench, We should not skip loading bloom filter blocks when they are not available in block cache when using Iterator::Seek
Actually I am not sure why this behavior existed in the first place
Closes https://github.com/facebook/rocksdb/pull/2255
Differential Revision: D5010721
Pulled By: maysamyabandeh
fbshipit-source-id: 0af545a06ac4baeecb248706ec34d009c2480ca4
Summary:
Any non-raw-data dependent object must be destructed before the table
closes. There was a bug of not doing that for filter object. This patch
fixes the bug and adds a unit test to prevent such bugs in future.
Closes https://github.com/facebook/rocksdb/pull/2246
Differential Revision: D5001318
Pulled By: maysamyabandeh
fbshipit-source-id: 6d8772e58765485868094b92964da82ef9730b6d
Summary:
Now if we have iterate_upper_bound set, we continue read until get a key >= upper_bound. For a lot of cases that neighboring data blocks have a user key gap between them, our index key will be a user key in the middle to get a shorter size. For example, if we have blocks:
[a b c d][f g h]
Then the index key for the first block will be 'e'.
then if upper bound is any key between 'd' and 'e', for example, d1, d2, ..., d99999999999, we don't have to read the second block and also know that we have done our iteration by reaching the last key that smaller the upper bound already.
This diff can reduce RA in most cases.
Closes https://github.com/facebook/rocksdb/pull/2239
Differential Revision: D4990693
Pulled By: lightmark
fbshipit-source-id: ab30ea2e3c6edf3fddd5efed3c34fcf7739827ff
Summary:
Some filters such as partitioned filter have pointers to the table for which they are created. Therefore is they are stored in the block cache, the should be forcibly erased from block cache before closing the table, which would result into deleting the object. Otherwise the destructor will be called later when the cache is lazily erasing the object, which having the parent table no longer existent it could result into undefined behavior.
Update: there will be still cases the filter is not removed from the cache since the table has not kept a pointer to the cache handle to be able to forcibly release it later. We make sure that the filter destructor does not access the table pointer to get around such cases.
Closes https://github.com/facebook/rocksdb/pull/2207
Differential Revision: D4941591
Pulled By: maysamyabandeh
fbshipit-source-id: 56fbab2a11cf447e1aa67caa30b58d7bd7ce5bbd
Summary:
With row cache being enabled, table cache is doing a short circuit for reading data. This path needs to be updated to take advantage of pinnable slice. In the meanwhile we disabling pinning in this path.
Closes https://github.com/facebook/rocksdb/pull/2237
Differential Revision: D4982389
Pulled By: maysamyabandeh
fbshipit-source-id: 542630d0cf23cfb1f0c397da82e7053df7966591
Summary:
Replacement of #2147
The change was squashed due to a lot of conflicts.
Closes https://github.com/facebook/rocksdb/pull/2194
Differential Revision: D4929799
Pulled By: siying
fbshipit-source-id: 5cd49c254737a1d5ac13f3c035f128e86524c581
Summary:
prefetch some data from the end of the file for each compaction to reduce IO.
Closes https://github.com/facebook/rocksdb/pull/2149
Differential Revision: D4880576
Pulled By: lightmark
fbshipit-source-id: aa767cd1afc84c541837fbf1ad6c0d45b34d3932
Summary:
This is an effort to club all string related utility functions into one common place, in string_util, so that it is easier for everyone to know what string processing functions are available. Right now they seem to be spread out across multiple modules, like logging and options_helper.
Check the sub-commits for easier reviewing.
Closes https://github.com/facebook/rocksdb/pull/2094
Differential Revision: D4837730
Pulled By: sagar0
fbshipit-source-id: 344278a
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:
to void future bug that caused by the mix of userkey/internalkey
Closes https://github.com/facebook/rocksdb/pull/2084
Differential Revision: D4825889
Pulled By: lightmark
fbshipit-source-id: 28411db
Summary:
Allow the users to specify the target index partition size.
With this patch an index partition is cut before its estimated in-memory size goes above the configured value for metadata_block_size. The filter partitions are still cut right after an index partition is cut.
Closes https://github.com/facebook/rocksdb/pull/2041
Differential Revision: D4780216
Pulled By: maysamyabandeh
fbshipit-source-id: 95a0831
Summary:
Fixes#1961 which causes a segfault when filter_policy is nullptr and both
pin_l0_filter_and_index_blocks_in_cache/cache_index_and_filter_blocks
are set.
Closes https://github.com/facebook/rocksdb/pull/2029
Differential Revision: D4764862
Pulled By: maysamyabandeh
fbshipit-source-id: 05bd695
Summary:
need to consistently include "rocksdb/persistent_cache.h" to fix internal build
Closes https://github.com/facebook/rocksdb/pull/2034
Differential Revision: D4768101
Pulled By: ajkr
fbshipit-source-id: 2ecb07f
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
Summary:
This is the second split of this pull request: https://github.com/facebook/rocksdb/pull/1891 which includes only the builder part. The testing will be included in the third split, where the reader is also included.
Closes https://github.com/facebook/rocksdb/pull/1952
Differential Revision: D4660272
Pulled By: maysamyabandeh
fbshipit-source-id: 36b3cf0
Summary:
This option is needed to be enabled for Direct IO
and I cannot think of a reason where we need to disable it
remove it and default it to true
Closes https://github.com/facebook/rocksdb/pull/1944
Differential Revision: D4641088
Pulled By: IslamAbdelRahman
fbshipit-source-id: d7085b9
Summary:
The assertion in Abandon() fails when called after Finish() fails. Finish() already closes the builder so there's no need to call Abandon().
Closes https://github.com/facebook/rocksdb/pull/1901
Differential Revision: D4601373
Pulled By: ajkr
fbshipit-source-id: e5678be
Summary:
Remove disableDataSync, and another similarly named disable_data_sync options.
This is being done to simplify options, and also because the performance gains of this feature can be achieved by other methods.
Closes https://github.com/facebook/rocksdb/pull/1859
Differential Revision: D4541292
Pulled By: sagar0
fbshipit-source-id: 5b3a6ca
Summary:
Partition Index blocks and use a Partition-index as a 2nd level index.
The two-level index can be used by setting
BlockBasedTableOptions::kTwoLevelIndexSearch as the index type and
configuring BlockBasedTableOptions::index_per_partition
t15539501
Closes https://github.com/facebook/rocksdb/pull/1814
Differential Revision: D4473535
Pulled By: maysamyabandeh
fbshipit-source-id: bffb87e
Summary:
introduce new methods into a public threadpool interface,
- allow submission of std::functions as they allow greater flexibility.
- add Joining methods to the implementation to join scheduled and submitted jobs with
an option to cancel jobs that did not start executing.
- Remove ugly `#ifdefs` between pthread and std implementation, make it uniform.
- introduce pimpl for a drop in replacement of the implementation
- Introduce rocksdb::port::Thread typedef which is a replacement for std::thread. On Posix Thread defaults as before std::thread.
- Implement WindowsThread that allocates memory in a more controllable manner than windows std::thread with a replaceable implementation.
- should be no functionality changes.
Closes https://github.com/facebook/rocksdb/pull/1823
Differential Revision: D4492902
Pulled By: siying
fbshipit-source-id: c74cb11
Summary:
merger.h was always a confusing name for me, simply give the file a better name
Closes https://github.com/facebook/rocksdb/pull/1836
Differential Revision: D4505357
Pulled By: IslamAbdelRahman
fbshipit-source-id: 07b28d8
Summary:
I added the Cache::Ref() function a couple weeks ago (#1761) to make this feature possible. Like other meta-blocks, rep_->range_del_entry holds a cache handle to pin the range deletion block in uncompressed block cache for the duration of the table reader's lifetime. We can reuse this cache handle to create an iterator over this meta-block without any cache lookup. Ref() is used to increment the cache handle's refcount in case the returned iterator outlives the table reader.
Closes https://github.com/facebook/rocksdb/pull/1801
Differential Revision: D4458782
Pulled By: ajkr
fbshipit-source-id: 2883f10
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:
1. value 100 byte: 1.8% regular, 1.2% merge values
2. value 1k byte: 11.5% regular, 7.5% merge values
3. 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 difference
is a little and could be noise. More importantly it is safely
cancelled
Closes https://github.com/facebook/rocksdb/pull/1732
Differential Revision: D4374613
Pulled By: maysamyabandeh
fbshipit-source-id: a077f1a
Summary:
Cleanable objects will perform the registered cleanups when
they are destructed. We however rather to delay this cleaning like when
we are gathering the merge operands. Current approach is to create the
Cleanable object on heap (instead of on stack) and delay deleting it.
By allowing Cleanables to delegate their cleanups to another cleanable
object we can delay the cleaning without however the need to craete the
cleanable object on heap and keeping it around. This patch applies this
technique for the cleanups of BlockIter and shows improved performance
for some in-memory benchmarks:
+1.8% for merge worklaod, +6.4% for non-merge workload when the merge
operator is specified.
https://our.intern.facebook.com/intern/tasks?t=15168163
Non-merge benchmark:
TEST_TMPDIR=/dev/shm/v100nocomp/ ./db_bench --benchmarks=fillrandom
--num=1000000 -value_size=100 -compression_type=none
Reading random with no merge operator specified:
TEST_TMPDIR=/dev/shm/v100nocomp/ ./db_bench
--benchmarks="read
Closes https://github.com/facebook/rocksdb/pull/1711
Differential Revision: D4361163
Pulled By: maysamyabandeh
fbshipit-source-id: 9801e07
Summary:
Add `fadvise_trigger` option to `SstFileWriter`
If fadvise_trigger is passed with a non-zero value, SstFileWriter will invalidate the os page cache every `fadvise_trigger` bytes for the sst file
Closes https://github.com/facebook/rocksdb/pull/1731
Differential Revision: D4371246
Pulled By: IslamAbdelRahman
fbshipit-source-id: 91caff1
Summary:
Improve cache options logging to info log.
Also print the value of
cache_index_and_filter_blocks_with_high_priority.
Closes https://github.com/facebook/rocksdb/pull/1709
Differential Revision: D4358776
Pulled By: yiwu-arbug
fbshipit-source-id: 8f030a0
Summary:
add a new function to SstFileWriter that will tell the user how big is there file right now.
Closes https://github.com/facebook/rocksdb/pull/1686
Differential Revision: D4338868
Pulled By: mdyuki1016
fbshipit-source-id: c1ee16a
Summary:
The gcc-7 code for parsing comments (libcpp/lex.c) didn't match
the intentional fallthough in this comment.
table/block_based_table_builder.cc: In member function 'void rocksdb::BlockBasedTableBuilder::WriteRawBlock(const rocksdb::Slice&, rocksdb::CompressionType, rocksdb::BlockHandle*)':
table/block_based_table_builder.cc:754:22: error: this statement may fall through [-Werror=implicit-fallthrough=]
assert(false);
^
table/block_based_table_builder.cc:756:7: note: here
case kCRC32c: {
^~~~
cc1plus: all warnings being treated as errors
Closes https://github.com/facebook/rocksdb/pull/1661
Differential Revision: D4318817
Pulled By: yiwu-arbug
fbshipit-source-id: e67d171
Summary:
Allow user to explicitly specify that the generated file by SstFileWriter will be ingested in a specific CF.
This allow us to persist the CF id in the generated file
Closes https://github.com/facebook/rocksdb/pull/1615
Differential Revision: D4270422
Pulled By: IslamAbdelRahman
fbshipit-source-id: 7fb954e
Summary:
- Made RangeDelAggregator's InternalKeyComparator member a reference-to-const so we don't need to copy-construct it. Also added InternalKeyComparator to ImmutableCFOptions so we don't need to construct one for each DBIter.
- Made MemTable::NewRangeTombstoneIterator and the table readers' NewRangeTombstoneIterator() functions return nullptr instead of NewEmptyInternalIterator to avoid the allocation. Updated callers accordingly.
Closes https://github.com/facebook/rocksdb/pull/1548
Differential Revision: D4208169
Pulled By: ajkr
fbshipit-source-id: 2fd65cf
Summary:
The Arena construction/destruction introduced significant overhead to read-heavy workload just by creating empty vectors for its blocks, so avoid it in RangeDelAggregator.
Closes https://github.com/facebook/rocksdb/pull/1547
Differential Revision: D4207781
Pulled By: ajkr
fbshipit-source-id: 9d1c130
Summary:
Previously we used TableCache::NewIterator() for multiple purposes (data
block iterator and range deletion iterator), and returned non-ok status in
the data block iterator. In one case where the caller only used the range
deletion block iterator (9e7cf3469b/db/version_set.cc (L965-L973)),
we didn't check/free the data block iterator containing non-ok status, which
caused a valgrind error.
So, this diff decouples creation of data block and range deletion block iterators,
and updates the callers accordingly. Both functions can return non-ok status
in an InternalIterator. Since the non-ok status is returned in an iterator that the
callers will definitely use, it should be more usable/less error-prone.
Closes https://github.com/facebook/rocksdb/pull/1513
Differential Revision: D4181423
Pulled By: ajkr
fbshipit-source-id: 835b8f5
Summary:
If user did not call SstFileWriter::Finish() or called Finish() but it failed.
We need to abandon the builder, to avoid destructing it while it's open
Closes https://github.com/facebook/rocksdb/pull/1502
Differential Revision: D4171660
Pulled By: IslamAbdelRahman
fbshipit-source-id: ab6f434
Summary:
Change DumpTable() so we can see the range deletion meta-block.
Closes https://github.com/facebook/rocksdb/pull/1505
Differential Revision: D4172227
Pulled By: ajkr
fbshipit-source-id: ae35665
Summary:
This handles two issues: (1) range deletion iterator sometimes outlives
the table reader that created it, in which case the block must not be destroyed
during table reader destruction; and (2) we prefer to read these range tombstone
meta-blocks from file fewer times.
- Extracted cache-populating logic from NewDataBlockIterator() into a separate function: MaybeLoadDataBlockToCache()
- Use MaybeLoadDataBlockToCache() to load range deletion meta-block and pin it through the reader's lifetime. This code reuse works since range deletion meta-block has same format as data blocks.
- Use NewDataBlockIterator() to create range deletion iterators, which uses block cache if enabled, otherwise reads the block from file. Either way, the underlying block won't disappear until after the iterator is destroyed.
Closes https://github.com/facebook/rocksdb/pull/1459
Differential Revision: D4123175
Pulled By: ajkr
fbshipit-source-id: 8f64281
Summary:
We can support SST files >2GB but we don't support blocks >2GB
Closes https://github.com/facebook/rocksdb/pull/1465
Differential Revision: D4132140
Pulled By: yiwu-arbug
fbshipit-source-id: 63bf12d
Summary:
During Get()/MultiGet(), build up a RangeDelAggregator with range
tombstones as we search through live memtable, immutable memtables, and
SST files. This aggregator is then used by memtable.cc's SaveValue() and
GetContext::SaveValue() to check whether keys are covered.
added tests for Get on memtables/files; end-to-end tests mainly in https://reviews.facebook.net/D64761
Closes https://github.com/facebook/rocksdb/pull/1456
Differential Revision: D4111271
Pulled By: ajkr
fbshipit-source-id: 6e388d4
Summary:
Changed BuildTable() (used for flush) to (1) add range
tombstones to the aggregator, which is used by CompactionIterator to
determine which keys can be removed; and (2) add aggregator's range
tombstones to the table that is output for the flush.
Closes https://github.com/facebook/rocksdb/pull/1438
Differential Revision: D4100025
Pulled By: ajkr
fbshipit-source-id: cb01a70
Summary: Siying suggested to keep old code for normal mode prev() for safety
Test Plan: make check -j64
Reviewers: yiwu, andrewkr, sdong
Reviewed By: sdong
Subscribers: andrewkr, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D65439
Summary:
change ioptions.comparator to user_comparator instread of internal_comparator.
Also change Comparator* to InternalKeyComparator* to make its type explicitly.
Test Plan: make all check -j64
Reviewers: andrewkr, sdong, yiwu
Reviewed By: yiwu
Subscribers: andrewkr, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D65121
Summary:
reland https://reviews.facebook.net/D62523
- Update SstFileWriter to include a property for a global sequence number in the SST file `rocksdb.external_sst_file.global_seqno`
- Update TableProperties to be aware of the offset of each property in the file
- Update BlockBasedTableReader and Block to be able to honor the sequence number in `rocksdb.external_sst_file.global_seqno` property and use it to overwrite all sequence number in the file
Something worth mentioning is that we don't update the seqno in the index block since and when doing a binary search, the reason for that is that it's guaranteed that SST files with global seqno will have only one user_key and each key will have seqno=0 encoded in it, This mean that this key is greater than any other key with seqno> 0. That mean that we can actually keep the current logic for these blocks
Test Plan: unit tests
Reviewers: sdong, yhchiang
Subscribers: andrewkr, dhruba
Differential Revision: https://reviews.facebook.net/D65211
Summary:
This diff introduces RangeDelAggregator, which takes ownership of iterators
provided to it via AddTombstones(). The tombstones are organized in a two-level
map (snapshot stripe -> begin key -> tombstone). Tombstone creation avoids data
copy by holding Slices returned by the iterator, which remain valid thanks to pinning.
For compaction, we create a hierarchical range tombstone iterator with structure
matching the iterator over compaction input data. An aggregator based on that
iterator is used by CompactionIterator to determine which keys are covered by
range tombstones. In case of merge operand, the same aggregator is used by
MergeHelper. Upon finishing each file in the compaction, relevant range tombstones
are added to the output file's range tombstone metablock and file boundaries are
updated accordingly.
To check whether a key is covered by range tombstone, RangeDelAggregator::ShouldDelete()
considers tombstones in the key's snapshot stripe. When this function is used outside of
compaction, it also checks newer stripes, which can contain covering tombstones. Currently
the intra-stripe check involves a linear scan; however, in the future we plan to collapse ranges
within a stripe such that binary search can be used.
RangeDelAggregator::AddToBuilder() adds all range tombstones in the table's key-range
to a new table's range tombstone meta-block. Since range tombstones may fall in the gap
between files, we may need to extend some files' key-ranges. The strategy is (1) first file
extends as far left as possible and other files do not extend left, (2) all files extend right
until either the start of the next file or the end of the last range tombstone in the gap,
whichever comes first.
One other notable change is adding release/move semantics to ScopedArenaIterator
such that it can be used to transfer ownership of an arena-allocated iterator, similar to
how unique_ptr is used for malloc'd data.
Depends on D61473
Test Plan: compaction_iterator_test, mock_table, end-to-end tests in D63927
Reviewers: sdong, IslamAbdelRahman, wanning, yhchiang, lightmark
Reviewed By: lightmark
Subscribers: andrewkr, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D62205
Summary:
1) The previous solution for Prev() prefix support is not clean.
Since I add api SeekForPrev(), now the Prev() can be symmetric to Next().
and we do not need SeekToLast() to be called in Prev() any more.
Also, Next() will Seek(prefix_seek_key_) to solve the problem of possible inconsistency between db_iter and merge_iter when
there is merge_operator. And prefix_seek_key is only refreshed when change direction to forward.
2) This diff also solves the bug of Iterator::SeekToLast() with iterate_upper_bound_ with prefix extractor.
add test cases for the above two cases.
There are some tests for the SeekToLast() in Prev(), I will clean them later.
Test Plan: make all check
Reviewers: IslamAbdelRahman, andrewkr, yiwu, sdong
Reviewed By: sdong
Subscribers: andrewkr, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D63933
Summary:
- Update SstFileWriter to include a property for a global sequence number in the SST file `rocksdb.external_sst_file.global_seqno`
- Update TableProperties to be aware of the offset of each property in the file
- Update BlockBasedTableReader and Block to be able to honor the sequence number in `rocksdb.external_sst_file.global_seqno` property and use it to overwrite all sequence number in the file
Something worth mentioning is that we don't update the seqno in the index block since and when doing a binary search, the reason for that is that it's guaranteed that SST files with global seqno will have only one user_key and each key will have seqno=0 encoded in it, This mean that this key is greater than any other key with seqno> 0. That mean that we can actually keep the current logic for these blocks
Test Plan: unit tests
Reviewers: andrewkr, yhchiang, yiwu, sdong
Reviewed By: sdong
Subscribers: hcz, andrewkr, dhruba
Differential Revision: https://reviews.facebook.net/D62523
Summary:
- Store range tombstones in a separate MemTableRep instantiated with ColumnFamilyOptions::memtable_factory
- MemTable::NewRangeTombstoneIterator() returns a MemTableIterator over the separate MemTableRep
- Part of the read path is not implemented yet (i.e., MemTable::Get())
Test Plan: see unit tests
Reviewers: wanning
Subscribers: andrewkr, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D62217
Summary:
Add new Iterator API, `SeekForPrev`: find the last key that <= target key
support prefix_extractor
support prefix_same_as_start
support upper_bound
not supported in iterators without Prev()
Also add tests in db_iter_test and db_iterator_test
Pass all tests
Cheers!
Test Plan: make all check -j64
Reviewers: andrewkr, yiwu, IslamAbdelRahman, sdong
Reviewed By: sdong
Subscribers: andrewkr, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D64149
