Summary:
The filter bits builder collects all the hashes to add in memory before adding them (because the number of keys is not known until we've walked over all the keys). Existing code uses a std::vector for this, which can mean up to 2x than necessary space allocated (and not freed) and up to ~2x write amplification in memory. Using std::deque uses close to minimal space (for large filters, the only time it matters), no write amplification, frees memory while building, and no need for large contiguous memory area. The only cost is more calls to allocator, which does not appear to matter, at least in benchmark test.
For now, this change only applies to the new (format_version=5) Bloom filter implementation, to ease before-and-after comparison downstream.
Temporary memory use during build is about the only way the new Bloom filter could regress vs. the old (because of upgrade to 64-bit hash) and that should only matter for full filters. This change should largely mitigate that potential regression.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/6175
Test Plan:
Using filter_bench with -new_builder option and 6M keys per filter is like large full filter (improvement). 10k keys and no -new_builder is like partitioned filters (about the same). (Corresponding configurations run simultaneously on devserver.)
std::vector impl (before)
$ /usr/bin/time -v ./filter_bench -impl=2 -quick -new_builder -working_mem_size_mb=1000 -
average_keys_per_filter=6000000
Build avg ns/key: 52.2027
Maximum resident set size (kbytes): 1105016
$ /usr/bin/time -v ./filter_bench -impl=2 -quick -working_mem_size_mb=1000 -
average_keys_per_filter=10000
Build avg ns/key: 30.5694
Maximum resident set size (kbytes): 1208152
std::deque impl (after)
$ /usr/bin/time -v ./filter_bench -impl=2 -quick -new_builder -working_mem_size_mb=1000 -
average_keys_per_filter=6000000
Build avg ns/key: 39.0697
Maximum resident set size (kbytes): 1087196
$ /usr/bin/time -v ./filter_bench -impl=2 -quick -working_mem_size_mb=1000 -
average_keys_per_filter=10000
Build avg ns/key: 30.9348
Maximum resident set size (kbytes): 1207980
Differential Revision: D19053431
Pulled By: pdillinger
fbshipit-source-id: 2888e748723a19d9ea40403934f13cbb8483430c
Summary:
The current Env API encompasses both storage/file operations, as well as OS related operations. Most of the APIs return a Status, which does not have enough metadata about an error, such as whether its retry-able or not, scope (i.e fault domain) of the error etc., that may be required in order to properly handle a storage error. The file APIs also do not provide enough control over the IO SLA, such as timeout, prioritization, hinting about placement and redundancy etc.
This PR separates out the file/storage APIs from Env into a new FileSystem class. The APIs are updated to return an IOStatus with metadata about the error, as well as to take an IOOptions structure as input in order to allow more control over the IO.
The user can set both ```options.env``` and ```options.file_system``` to specify that RocksDB should use the former for OS related operations and the latter for storage operations. Internally, a ```CompositeEnvWrapper``` has been introduced that inherits from ```Env``` and redirects individual methods to either an ```Env``` implementation or the ```FileSystem``` as appropriate. When options are sanitized during ```DB::Open```, ```options.env``` is replaced with a newly allocated ```CompositeEnvWrapper``` instance if both env and file_system have been specified. This way, the rest of the RocksDB code can continue to function as before.
This PR also ports PosixEnv to the new API by splitting it into two - PosixEnv and PosixFileSystem. PosixEnv is defined as a sub-class of CompositeEnvWrapper, and threading/time functions are overridden with Posix specific implementations in order to avoid an extra level of indirection.
The ```CompositeEnvWrapper``` translates ```IOStatus``` return code to ```Status```, and sets the severity to ```kSoftError``` if the io_status is retryable. The error handling code in RocksDB can then recover the DB automatically.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5761
Differential Revision: D18868376
Pulled By: anand1976
fbshipit-source-id: 39efe18a162ea746fabac6360ff529baba48486f
Summary:
And clean up related code, especially in stress test.
(More clean up of db_stress_test_base.cc coming after this.)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/6154
Test Plan: make check, make blackbox_crash_test for a bit
Differential Revision: D18938180
Pulled By: pdillinger
fbshipit-source-id: 524d27621b8dbb25f6dff40f1081e7c00630357e
Summary:
This change fixes a source issue that caused compile time error which breaks build for many fbcode services in that setup. The size() member function of channel is a const member, so member variables accessed within it are implicitly const as well. This caused error when clang fails to resolve to a constructor that takes std::mutex because the suitable constructor got rejected due to loss of constness for its argument. The fix is to add mutable modifier to the lock_ member of channel.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/6161
Differential Revision: D18967685
Pulled By: maysamyabandeh
fbshipit-source-id: 698b6a5153c3c92eeacb842c467aa28cc350d432
Summary:
thread_local_test now fails because it asserts no thread local instance is created when the test started. However, right now a thread local instance might be created when creating PosixEnv as a static variable. Fix the test by relaxing the assumption of starting from 0.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/6136
Test Plan: Find an environment where the test fails, and see it passes with the fix applied.
Differential Revision: D18889224
fbshipit-source-id: 7946f3bfea81d236f7bb1554076696705b211b92
Summary:
From the reset of the code, it looks this this maybe can be unconditionally given the attribute? But I couldn't test with MSVC so I defensively put under CPP.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/6075
Differential Revision: D18723749
fbshipit-source-id: 45fc8732c28dd29aab1644225d68f3c6f39bd69b
Summary:
This change enables custom implementations of FilterPolicy to
wrap a variety of NewBloomFilterPolicy and select among them based on
contextual information such as table level and compaction style.
* Moves FilterBuildingContext to public API and elaborates it with more
useful data. (It would be nice to put more general options-like data,
but at the time this object is constructed, we are using internal APIs
ImmutableCFOptions and MutableCFOptions and don't have easy access to
ColumnFamilyOptions that I can tell.)
* Renames BloomFilterPolicy::GetFilterBitsBuilderInternal to
GetBuilderWithContext, because it's now public.
* Plumbs through the table's "level_at_creation" for filter building
context.
* Simplified some tests by adding GetBuilder() to
MockBlockBasedTableTester.
* Adds test as DBBloomFilterTest.ContextCustomFilterPolicy, including
sample wrapper class LevelAndStyleCustomFilterPolicy.
* Fixes a cross-test bug in DBBloomFilterTest.OptimizeFiltersForHits
where it does not reset perf context.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/6088
Test Plan: make check, valgrind on db_bloom_filter_test
Differential Revision: D18697817
Pulled By: pdillinger
fbshipit-source-id: 5f987a2d7b07cc7a33670bc08ca6b4ca698c1cf4
Summary:
There's no technological impediment to allowing the Bloom
filter bits/key to be non-integer (fractional/decimal) values, and it
provides finer control over the memory vs. accuracy trade-off. This is
especially handy in using the format_version=5 Bloom filter in place
of the old one, because bits_per_key=9.55 provides the same accuracy as
the old bits_per_key=10.
This change not only requires refining the logic for choosing the best
num_probes for a given bits/key setting, it revealed a flaw in that logic.
As bits/key gets higher, the best num_probes for a cache-local Bloom
filter is closer to bpk / 2 than to bpk * 0.69, the best choice for a
standard Bloom filter. For example, at 16 bits per key, the best
num_probes is 9 (FP rate = 0.0843%) not 11 (FP rate = 0.0884%).
This change fixes and refines that logic (for the format_version=5
Bloom filter only, just in case) based on empirical tests to find
accuracy inflection points between each num_probes.
Although bits_per_key is now specified as a double, the new Bloom
filter converts/rounds this to "millibits / key" for predictable/precise
internal computations. Just in case of unforeseen compatibility
issues, we round to the nearest whole number bits / key for the
legacy Bloom filter, so as not to unlock new behaviors for it.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/6092
Test Plan: unit tests included
Differential Revision: D18711313
Pulled By: pdillinger
fbshipit-source-id: 1aa73295f152a995328cb846ef9157ae8a05522a
Summary:
As described in detail in issue https://github.com/facebook/rocksdb/issues/6048, iterators' dereference operators
(`*`, `->`, and `[]`) should return `pointer`s/`reference`s (as opposed to
`const_pointer`s/`const_reference`s) even if the iterator itself is `const`
to be in sync with the standard's iterator concept.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/6057
Test Plan: make check
Differential Revision: D18623235
Pulled By: ltamasi
fbshipit-source-id: 04e82d73bc0c67fb0ded018383af8dfc332050cc
Summary:
This is a required operator for random-access iterators, and an upcoming update for Visual Studio 2019 will change the C++ Standard Library's heap algorithms to use this operator.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/6047
Differential Revision: D18618531
Pulled By: ltamasi
fbshipit-source-id: 08d10bc85bf2dbc3f7ef0fa3c777e99f1e927ef5
Summary:
Adds an improved, replacement Bloom filter implementation (FastLocalBloom) for full and partitioned filters in the block-based table. This replacement is faster and more accurate, especially for high bits per key or millions of keys in a single filter.
Speed
The improved speed, at least on recent x86_64, comes from
* Using fastrange instead of modulo (%)
* Using our new hash function (XXH3 preview, added in a previous commit), which is much faster for large keys and only *slightly* slower on keys around 12 bytes if hashing the same size many thousands of times in a row.
* Optimizing the Bloom filter queries with AVX2 SIMD operations. (Added AVX2 to the USE_SSE=1 build.) Careful design was required to support (a) SIMD-optimized queries, (b) compatible non-SIMD code that's simple and efficient, (c) flexible choice of number of probes, and (d) essentially maximized accuracy for a cache-local Bloom filter. Probes are made eight at a time, so any number of probes up to 8 is the same speed, then up to 16, etc.
* Prefetching cache lines when building the filter. Although this optimization could be applied to the old structure as well, it seems to balance out the small added cost of accumulating 64 bit hashes for adding to the filter rather than 32 bit hashes.
Here's nominal speed data from filter_bench (200MB in filters, about 10k keys each, 10 bits filter data / key, 6 probes, avg key size 24 bytes, includes hashing time) on Skylake DE (relatively low clock speed):
$ ./filter_bench -quick -impl=2 -net_includes_hashing # New Bloom filter
Build avg ns/key: 47.7135
Mixed inside/outside queries...
Single filter net ns/op: 26.2825
Random filter net ns/op: 150.459
Average FP rate %: 0.954651
$ ./filter_bench -quick -impl=0 -net_includes_hashing # Old Bloom filter
Build avg ns/key: 47.2245
Mixed inside/outside queries...
Single filter net ns/op: 63.2978
Random filter net ns/op: 188.038
Average FP rate %: 1.13823
Similar build time but dramatically faster query times on hot data (63 ns to 26 ns), and somewhat faster on stale data (188 ns to 150 ns). Performance differences on batched and skewed query loads are between these extremes as expected.
The only other interesting thing about speed is "inside" (query key was added to filter) vs. "outside" (query key was not added to filter) query times. The non-SIMD implementations are substantially slower when most queries are "outside" vs. "inside". This goes against what one might expect or would have observed years ago, as "outside" queries only need about two probes on average, due to short-circuiting, while "inside" always have num_probes (say 6). The problem is probably the nastily unpredictable branch. The SIMD implementation has few branches (very predictable) and has pretty consistent running time regardless of query outcome.
Accuracy
The generally improved accuracy (re: Issue https://github.com/facebook/rocksdb/issues/5857) comes from a better design for probing indices
within a cache line (re: Issue https://github.com/facebook/rocksdb/issues/4120) and improved accuracy for millions of keys in a single filter from using a 64-bit hash function (XXH3p). Design details in code comments.
Accuracy data (generalizes, except old impl gets worse with millions of keys):
Memory bits per key: FP rate percent old impl -> FP rate percent new impl
6: 5.70953 -> 5.69888
8: 2.45766 -> 2.29709
10: 1.13977 -> 0.959254
12: 0.662498 -> 0.411593
16: 0.353023 -> 0.0873754
24: 0.261552 -> 0.0060971
50: 0.225453 -> ~0.00003 (less than 1 in a million queries are FP)
Fixes https://github.com/facebook/rocksdb/issues/5857
Fixes https://github.com/facebook/rocksdb/issues/4120
Unlike the old implementation, this implementation has a fixed cache line size (64 bytes). At 10 bits per key, the accuracy of this new implementation is very close to the old implementation with 128-byte cache line size. If there's sufficient demand, this implementation could be generalized.
Compatibility
Although old releases would see the new structure as corrupt filter data and read the table as if there's no filter, we've decided only to enable the new Bloom filter with new format_version=5. This provides a smooth path for automatic adoption over time, with an option for early opt-in.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/6007
Test Plan: filter_bench has been used thoroughly to validate speed, accuracy, and correctness. Unit tests have been carefully updated to exercise new and old implementations, as well as the logic to select an implementation based on context (format_version).
Differential Revision: D18294749
Pulled By: pdillinger
fbshipit-source-id: d44c9db3696e4d0a17caaec47075b7755c262c5f
Summary:
For upcoming new SST filter implementations, we will use a new
64-bit hash function (XXH3 preview, slightly modified). This change
updates hash.{h,cc} for that change, adds unit tests, and out-of-lines
the implementations to keep hash.h as clean/small as possible.
In developing the unit tests, I discovered that the XXH3 preview always
returns zero for the empty string. Zero is problematic for some
algorithms (including an upcoming SST filter implementation) if it
occurs more often than at the "natural" rate, so it should not be
returned from trivial values using trivial seeds. I modified our fork
of XXH3 to return a modest hash of the seed for the empty string.
With hash function details out-of-lines in hash.h, it makes sense to
enable XXH_INLINE_ALL, so that direct calls to XXH64/XXH32/XXH3p
are inlined. To fix array-bounds warnings on some inline calls, I
injected some casts to uintptr_t in xxhash.cc. (Issue reported to Yann.)
Revised: Reverted using XXH_INLINE_ALL for now. Some Facebook
checks are unhappy about #include on xxhash.cc file. I would
fix that by rename to xxhash_cc.h, but to best preserve history I want
to do that in a separate commit (PR) from the uintptr casts.
Also updated filter_bench for this change, improving the performance
predictability of dry run hashing and adding support for 64-bit hash
(for upcoming new SST filter implementations, minor dead code in the
tool for now).
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5984
Differential Revision: D18246567
Pulled By: pdillinger
fbshipit-source-id: 6162fbf6381d63c8cc611dd7ec70e1ddc883fbb8
Summary:
This change sets up for alternate implementations underlying
BloomFilterPolicy:
* Refactor BloomFilterPolicy and expose in internal .h file so that it's easy to iterate over / select implementations for testing, regardless of what the best public interface will look like. Most notably updated db_bloom_filter_test to use this.
* Hide FullFilterBitsBuilder from unit tests (alternate derived classes planned); expose the part important for testing (CalculateSpace), as abstract class BuiltinFilterBitsBuilder. (Also cleaned up internally exposed interface to CalculateSpace.)
* Rename BloomTest -> BlockBasedBloomTest for clarity (despite ongoing confusion between block-based table and block-based filter)
* Assert that block-based filter construction interface is only used on BloomFilterPolicy appropriately constructed. (A couple of tests updated to add ", true".)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5967
Test Plan: make check
Differential Revision: D18138704
Pulled By: pdillinger
fbshipit-source-id: 55ef9273423b0696309e251f50b8c1b5e9ec7597
Summary:
filter_bench is a specialized micro-benchmarking tool that
should not be needed with ROCKSDB_LITE. This should fix the LITE build.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5978
Test Plan: make LITE=1 check
Differential Revision: D18177941
Pulled By: pdillinger
fbshipit-source-id: b73a171404661e09e018bc99afcf8d4bf1e2949c
Summary:
* Adds support for plain table filter. This is not critical right now, but does add a -impl flag that will be useful for new filter implementations initially targeted at block-based table (and maybe later ported to plain table)
* Better mixing of inside vs. outside queries, for more realism
* A -best_case option handy for implementation tuning inner loop
* Option for whether to include hashing time in dry run / net timings
No modifications to production code, just filter_bench.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5968
Differential Revision: D18139872
Pulled By: pdillinger
fbshipit-source-id: 5b09eba963111b48f9e0525a706e9921070990e8
Summary:
Some filtering tests were unfriendly to new implementations of
FilterBitsBuilder because of dynamic_cast to FullFilterBitsBuilder. Most
of those have now been cleaned up, worked around, or at least changed
from crash on dynamic_cast failure to individual test failure.
Also put some clarifying comments on filter-related APIs.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5960
Test Plan: make check
Differential Revision: D18121223
Pulled By: pdillinger
fbshipit-source-id: e83827d9d5d96315d96f8e25a99cd70f497d802c
Summary:
- Updated our included xxhash implementation to version 0.7.2 (== the latest dev version as of 2019-10-09).
- Using XXH_NAMESPACE (like other fb projects) to avoid potential name collisions.
- Added fastrange64, and unit tests for it and fastrange32. These are faster alternatives to hash % range.
- Use preview version of XXH3 instead of MurmurHash64A for NPHash64
-- Had to update cache_test to increase probability of passing for any given hash function.
- Use fastrange64 instead of % with uses of NPHash64
-- Had to fix WritePreparedTransactionTest.CommitOfDelayedPrepared to avoid deadlock apparently caused by new hash collision.
- Set default seed for NPHash64 because specifying a seed rarely makes sense for it.
- Removed unnecessary include xxhash.h in a popular .h file
- Rename preview version of XXH3 to XXH3p for clarity and to ease backward compatibility in case final version of XXH3 is integrated.
Relying on existing unit tests for NPHash64-related changes. Each new implementation of fastrange64 passed unit tests when manipulating my local build to select it. I haven't done any integration performance tests, but I consider the improved performance of the pieces being swapped in to be well established.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5909
Differential Revision: D18125196
Pulled By: pdillinger
fbshipit-source-id: f6bf83d49d20cbb2549926adf454fd035f0ecc0d
Summary:
The parts that are used to implement FilterPolicy /
NewBloomFilterPolicy and not used other than for the block-based table
should be consolidated under table/block_based/filter_policy*.
This change is step 2 of 2:
mv util/bloom.cc table/block_based/filter_policy.cc
This gets its own PR so that git has the best chance of following the
rename for blame purposes. Note that low-level shared implementation
details of Bloom filters remain in util/bloom_impl.h, and
util/bloom_test.cc remains where it is for now.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5966
Test Plan: make check
Differential Revision: D18124930
Pulled By: pdillinger
fbshipit-source-id: 823bc09025b3395f092ef46a46aa5ba92a914d84
Summary:
The parts that are used to implement FilterPolicy /
NewBloomFilterPolicy and not used other than for the block-based table
should be consolidated under table/block_based/filter_policy*. I don't
foresee sharing these APIs with e.g. the Plain Table because they don't
expose hashes for reuse in indexing.
This change is step 1 of 2:
(a) mv table/full_filter_bits_builder.h to
table/block_based/filter_policy_internal.h which I expect to expand
soon to internally reveal more implementation details for testing.
(b) consolidate eventual contents of table/block_based/filter_policy.cc
in util/bloom.cc, which has the most elaborate revision history
(see step 2 ...)
Step 2 soon to follow:
mv util/bloom.cc table/block_based/filter_policy.cc
This gets its own PR so that git has the best chance of following the
rename for blame purposes. Note that low-level shared implementation
details of Bloom filters are in util/bloom_impl.h.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5963
Test Plan: make check
Differential Revision: D18121199
Pulled By: pdillinger
fbshipit-source-id: 8f21732c3d8909777e3240e4ac3123d73140326a
Summary:
The first version of filter_bench has selectable key size
but that size does not vary throughout a test run. This artificially
favors "branchy" hash functions like the existing BloomHash,
MurmurHash1, probably because of optimal return for branch prediction.
This change primarily varies those key sizes from -2 to +2 bytes vs.
the average selected size. We also set the default key size at 24 to
better reflect our best guess of typical key size.
But steadily random key sizes may not be realistic either. So this
change introduces a new filter_bench option:
-vary_key_size_log2_interval=n where the same key size is used 2^n
times and then changes to another size. I've set the default at 5
(32 times same size) as a compromise between deployments with
rather consistent vs. rather variable key sizes. On my Skylake
system, the performance boost to MurmurHash1 largely lies between
n=10 and n=15.
Also added -vary_key_alignment (bool, now default=true), though this
doesn't currently seem to matter in hash functions under
consideration.
This change also does a "dry run" for each testing scenario, to improve
the accuracy of those numbers, as there was more difference between
scenarios than expected. Subtracting gross test run times from dry run
times is now also embedded in the output, because these "net" times are
generally the most useful.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5933
Differential Revision: D18121683
Pulled By: pdillinger
fbshipit-source-id: 3c7efee1c5661a5fe43de555e786754ddf80dc1e
Summary:
This is an internal, file-local "feature" that is not used and
potentially confusing.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5961
Test Plan: make check
Differential Revision: D18099018
Pulled By: pdillinger
fbshipit-source-id: 7870627eeed09941d12538ec55d10d2e164fc716
Summary:
Amongst other things, PR https://github.com/facebook/rocksdb/issues/5504 refactored the filter block readers so that
only the filter block contents are stored in the block cache (as opposed to the
earlier design where the cache stored the filter block reader itself, leading to
potentially dangling pointers and concurrency bugs). However, this change
introduced a performance hit since with the new code, the metadata fields are
re-parsed upon every access. This patch reunites the block contents with the
filter bits reader to eliminate this overhead; since this is still a self-contained
pure data object, it is safe to store it in the cache. (Note: this is similar to how
the zstd digest is handled.)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5936
Test Plan:
make asan_check
filter_bench results for the old code:
```
$ ./filter_bench -quick
WARNING: Assertions are enabled; benchmarks unnecessarily slow
Building...
Build avg ns/key: 26.7153
Number of filters: 16669
Total memory (MB): 200.009
Bits/key actual: 10.0647
----------------------------
Inside queries...
Dry run (46b) ns/op: 33.4258
Single filter ns/op: 42.5974
Random filter ns/op: 217.861
----------------------------
Outside queries...
Dry run (25d) ns/op: 32.4217
Single filter ns/op: 50.9855
Random filter ns/op: 219.167
Average FP rate %: 1.13993
----------------------------
Done. (For more info, run with -legend or -help.)
$ ./filter_bench -quick -use_full_block_reader
WARNING: Assertions are enabled; benchmarks unnecessarily slow
Building...
Build avg ns/key: 26.5172
Number of filters: 16669
Total memory (MB): 200.009
Bits/key actual: 10.0647
----------------------------
Inside queries...
Dry run (46b) ns/op: 32.3556
Single filter ns/op: 83.2239
Random filter ns/op: 370.676
----------------------------
Outside queries...
Dry run (25d) ns/op: 32.2265
Single filter ns/op: 93.5651
Random filter ns/op: 408.393
Average FP rate %: 1.13993
----------------------------
Done. (For more info, run with -legend or -help.)
```
With the new code:
```
$ ./filter_bench -quick
WARNING: Assertions are enabled; benchmarks unnecessarily slow
Building...
Build avg ns/key: 25.4285
Number of filters: 16669
Total memory (MB): 200.009
Bits/key actual: 10.0647
----------------------------
Inside queries...
Dry run (46b) ns/op: 31.0594
Single filter ns/op: 43.8974
Random filter ns/op: 226.075
----------------------------
Outside queries...
Dry run (25d) ns/op: 31.0295
Single filter ns/op: 50.3824
Random filter ns/op: 226.805
Average FP rate %: 1.13993
----------------------------
Done. (For more info, run with -legend or -help.)
$ ./filter_bench -quick -use_full_block_reader
WARNING: Assertions are enabled; benchmarks unnecessarily slow
Building...
Build avg ns/key: 26.5308
Number of filters: 16669
Total memory (MB): 200.009
Bits/key actual: 10.0647
----------------------------
Inside queries...
Dry run (46b) ns/op: 33.2968
Single filter ns/op: 58.6163
Random filter ns/op: 291.434
----------------------------
Outside queries...
Dry run (25d) ns/op: 32.1839
Single filter ns/op: 66.9039
Random filter ns/op: 292.828
Average FP rate %: 1.13993
----------------------------
Done. (For more info, run with -legend or -help.)
```
Differential Revision: D17991712
Pulled By: ltamasi
fbshipit-source-id: 7ea205550217bfaaa1d5158ebd658e5832e60f29
Summary:
FullFilterBitsReader, after creating in BloomFilterPolicy, was
responsible for decoding metadata bits. This meant that
FullFilterBitsReader::MayMatch had some metadata checks in order to
implement "always true" or "always false" functionality in the case
of inconsistent or trivial metadata. This made for ugly
mixing-of-concerns code and probably had some runtime cost. It also
didn't really support plugging in alternative filter implementations
with extensions to the existing metadata schema.
BloomFilterPolicy::GetFilterBitsReader is now (exclusively) responsible
for decoding filter metadata bits and constructing appropriate instances
deriving from FilterBitsReader. "Always false" and "always true" derived
classes allow FullFilterBitsReader not to be concerned with handling of
trivial or inconsistent metadata. This also makes for easy expansion
to alternative filter implementations in new, alternative derived
classes. This change makes calls to FilterBitsReader::MayMatch
*necessarily* virtual because there's now more than one built-in
implementation. Compared with the previous implementation's extra
'if' checks in MayMatch, there's no consistent performance difference,
measured by (an older revision of) filter_bench (differences here seem
to be within noise):
Inside queries...
- Dry run (407) ns/op: 35.9996
+ Dry run (407) ns/op: 35.2034
- Single filter ns/op: 47.5483
+ Single filter ns/op: 47.4034
- Batched, prepared ns/op: 43.1559
+ Batched, prepared ns/op: 42.2923
...
- Random filter ns/op: 150.697
+ Random filter ns/op: 149.403
----------------------------
Outside queries...
- Dry run (980) ns/op: 34.6114
+ Dry run (980) ns/op: 34.0405
- Single filter ns/op: 56.8326
+ Single filter ns/op: 55.8414
- Batched, prepared ns/op: 48.2346
+ Batched, prepared ns/op: 47.5667
- Random filter ns/op: 155.377
+ Random filter ns/op: 153.942
Average FP rate %: 1.1386
Also, the FullFilterBitsReader ctor was responsible for a surprising
amount of CPU in production, due in part to inefficient determination of
the CACHE_LINE_SIZE used to construct the filter being read. The
overwhelming common case (same as my CACHE_LINE_SIZE) is now
substantially optimized, as shown with filter_bench with
-new_reader_every=1 (old option - see below) (repeatable result):
Inside queries...
- Dry run (453) ns/op: 118.799
+ Dry run (453) ns/op: 105.869
- Single filter ns/op: 82.5831
+ Single filter ns/op: 74.2509
...
- Random filter ns/op: 224.936
+ Random filter ns/op: 194.833
----------------------------
Outside queries...
- Dry run (aa1) ns/op: 118.503
+ Dry run (aa1) ns/op: 104.925
- Single filter ns/op: 90.3023
+ Single filter ns/op: 83.425
...
- Random filter ns/op: 220.455
+ Random filter ns/op: 175.7
Average FP rate %: 1.13886
However PR#5936 has/will reclaim most of this cost. After that PR, the optimization of this code path is likely negligible, but nonetheless it's clear we aren't making performance any worse.
Also fixed inadequate check of consistency between filter data size and
num_lines. (Unit test updated.)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5941
Test Plan:
previously added unit tests FullBloomTest.CorruptFilters and
FullBloomTest.RawSchema
Differential Revision: D18018353
Pulled By: pdillinger
fbshipit-source-id: 8e04c2b4a7d93223f49a237fd52ef2483929ed9c
Summary:
RocksDB has a MultiGet() API that implements batched key lookup for higher performance (https://github.com/facebook/rocksdb/blob/master/include/rocksdb/db.h#L468). Currently, batching is implemented in BlockBasedTableReader::MultiGet() for SST file lookups. One of the ways it improves performance is by pipelining bloom filter lookups (by prefetching required cachelines for all the keys in the batch, and then doing the probe) and thus hiding the cache miss latency. The same concept can be extended to the memtable as well. This PR involves implementing a pipelined bloom filter lookup in DynamicBloom, and implementing MemTable::MultiGet() that can leverage it.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5818
Test Plan:
Existing tests
Performance Test:
Ran the below command which fills up the memtable and makes sure there are no flushes and then call multiget. Ran it on master and on the new change and see atleast 1% performance improvement across all the test runs I did. Sometimes the improvement was upto 5%.
TEST_TMPDIR=/data/users/$USER/benchmarks/feature/ numactl -C 10 ./db_bench -benchmarks="fillseq,multireadrandom" -num=600000 -compression_type="none" -level_compaction_dynamic_level_bytes -write_buffer_size=200000000 -target_file_size_base=200000000 -max_bytes_for_level_base=16777216 -reads=90000 -threads=1 -compression_type=none -cache_size=4194304000 -batch_size=32 -disable_auto_compactions=true -bloom_bits=10 -cache_index_and_filter_blocks=true -pin_l0_filter_and_index_blocks_in_cache=true -multiread_batched=true -multiread_stride=4 -statistics -memtable_whole_key_filtering=true -memtable_bloom_size_ratio=10
Differential Revision: D17578869
Pulled By: vjnadimpalli
fbshipit-source-id: 23dc651d9bf49db11d22375bf435708875a1f192
Summary:
Fixed some spots where converting size_t or uint_fast32_t to
uint32_t. Wrapped mt19937 in a new Random32 class to avoid future
such traps.
NB: I tried using Random32::Uniform (std::uniform_int_distribution) in
filter_bench instead of fastrange, but that more than doubled the dry
run time! So I added fastrange as Random32::Uniformish. ;)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5894
Test Plan: USE_CLANG=1 build, and manual re-run filter_bench
Differential Revision: D17825131
Pulled By: pdillinger
fbshipit-source-id: 68feee333b5f8193c084ded760e3d6679b405ecd
Summary:
Example: using the tool before and after PR https://github.com/facebook/rocksdb/issues/5784 shows that
the refactoring, presumed performance-neutral, actually sped up SST
filters by about 3% to 8% (repeatable result):
Before:
- Dry run ns/op: 22.4725
- Single filter ns/op: 51.1078
- Random filter ns/op: 120.133
After:
+ Dry run ns/op: 22.2301
+ Single filter run ns/op: 47.4313
+ Random filter ns/op: 115.9
Only tests filters for the block-based table (full filters and
partitioned filters - same implementation; not block-based filters),
which seems to be the recommended format/implementation.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5825
Differential Revision: D17804987
Pulled By: pdillinger
fbshipit-source-id: 0f18a9c254c57f7866030d03e7fa4ba503bac3c5
Summary:
Broken type for shift in PR#5834. Fixing code means fixing
expected values in test.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5882
Test Plan: thisisthetest
Differential Revision: D17746136
Pulled By: pdillinger
fbshipit-source-id: d3c456ed30b433d55fcab6fc7d836940fe3b46b8
Summary:
There was significant untested logic in FullFilterBitsReader in
the handling of serialized Bloom filter bits that cannot be generated by
FullFilterBitsBuilder in the current compilation. These now test many of
those corner-case behaviors, including bad metadata or filters created
with different cache line size than the current compiled-in value.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5834
Test Plan: thisisthetest
Differential Revision: D17726372
Pulled By: pdillinger
fbshipit-source-id: fb7b8003b5a8e6fb4666fe95206128f3d5835fc7
Summary:
Further apply formatter to more recent commits.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5830
Test Plan: Run all existing tests.
Differential Revision: D17488031
fbshipit-source-id: 137458fd94d56dd271b8b40c522b03036943a2ab
Summary:
Some recent commits might not have passed through the formatter. I formatted recent 45 commits. The script hangs for more commits so I stopped there.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5827
Test Plan: Run all existing tests.
Differential Revision: D17483727
fbshipit-source-id: af23113ee63015d8a43d89a3bc2c1056189afe8f
Summary:
clang-analyzer has uncovered a bunch of places where the code is relying
on pointers being valid and one case (in VectorIterator) where a moved-from
object is being used:
In file included from db/range_tombstone_fragmenter.cc:17:
./util/vector_iterator.h:23:18: warning: Method called on moved-from object 'keys' of type 'std::vector'
current_(keys.size()) {
^~~~~~~~~~~
1 warning generated.
utilities/persistent_cache/block_cache_tier_file.cc:39:14: warning: Called C++ object pointer is null
Status s = env->NewRandomAccessFile(filepath, file, opt);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
utilities/persistent_cache/block_cache_tier_file.cc:47:19: warning: Called C++ object pointer is null
Status status = env_->GetFileSize(Path(), size);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
utilities/persistent_cache/block_cache_tier_file.cc:290:14: warning: Called C++ object pointer is null
Status s = env_->FileExists(Path());
^~~~~~~~~~~~~~~~~~~~~~~~
utilities/persistent_cache/block_cache_tier_file.cc:363:35: warning: Called C++ object pointer is null
CacheWriteBuffer* const buf = alloc_->Allocate();
^~~~~~~~~~~~~~~~~~
utilities/persistent_cache/block_cache_tier_file.cc:399:41: warning: Called C++ object pointer is null
const uint64_t file_off = buf_doff_ * alloc_->BufferSize();
^~~~~~~~~~~~~~~~~~~~
utilities/persistent_cache/block_cache_tier_file.cc:463:33: warning: Called C++ object pointer is null
size_t start_idx = lba.off_ / alloc_->BufferSize();
^~~~~~~~~~~~~~~~~~~~
utilities/persistent_cache/block_cache_tier_file.cc:515:5: warning: Called C++ object pointer is null
alloc_->Deallocate(bufs_[i]);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~
7 warnings generated.
ar: creating librocksdb_debug.a
utilities/memory/memory_test.cc:68:25: warning: Called C++ object pointer is null
cache_set->insert(db->GetDBOptions().row_cache.get());
^~~~~~~~~~~~~~~~~~
1 warning generated.
The patch fixes these by adding assertions and explicitly passing in zero
when initializing VectorIterator::current_ (which preserves the existing
behavior).
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5821
Test Plan: Ran make check and make analyze to make sure the warnings have disappeared.
Differential Revision: D17455949
Pulled By: ltamasi
fbshipit-source-id: 363619618ea649a0674287f9f3b3393e390571ee
Summary:
Manual compaction may bring in very high load because sometime the amount of data involved in a compaction could be large, which may affect online service. So it would be good if the running compaction making the server busy can be stopped immediately. In this implementation, stopping manual compaction condition is only checked in slow process. We let deletion compaction and trivial move go through.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/3971
Test Plan: add tests at more spots.
Differential Revision: D17369043
fbshipit-source-id: 575a624fb992ce0bb07d9443eb209e547740043c
Summary:
Refactoring to consolidate implementation details of legacy
Bloom filters. This helps to organize and document some related,
obscure code.
Also added make/cpp var TEST_CACHE_LINE_SIZE so that it's easy to
compile and run unit tests for non-native cache line size. (Fixed a
related test failure in db_properties_test.)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5784
Test Plan:
make check, including Recently added Bloom schema unit tests
(in ./plain_table_db_test && ./bloom_test), and including with
TEST_CACHE_LINE_SIZE=128U and TEST_CACHE_LINE_SIZE=256U. Tested the
schema tests with temporary fault injection into new implementations.
Some performance testing with modified unit tests suggest a small to moderate
improvement in speed.
Differential Revision: D17381384
Pulled By: pdillinger
fbshipit-source-id: ee42586da996798910fc45ac0b6289147f16d8df
Summary:
This will allow us to fix history by having the code changes for PR#5784 properly attributed to it.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5810
Differential Revision: D17400231
Pulled By: pdillinger
fbshipit-source-id: 2da8b1cdf2533cfedb35b5526eadefb38c291f09
Summary:
file_reader_writer.h and .cc contain several files and helper function, and it's hard to navigate. Separate it to multiple files and put them under file/
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5803
Test Plan: Build whole project using make and cmake.
Differential Revision: D17374550
fbshipit-source-id: 10efca907721e7a78ed25bbf74dc5410dea05987
Summary:
DynamicBloom unit test now tests non-sequential as well as
sequential keys in testing FP rates. Also now verifies larger structures.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5805
Test Plan: thisisthetest
Differential Revision: D17398109
Pulled By: pdillinger
fbshipit-source-id: 374074206c76d242efa378afc27830448a0e892a
Summary:
prefetch data for following block，avoid cache miss when doing crc caculate
I do performance test at kunpeng-920 server(arm-v8, 64core@2.6GHz)
./db_bench --benchmarks=crc32c --block_size=500000000
before optimise : 587313.500 micros/op 1 ops/sec; 811.9 MB/s (500000000 per op)
after optimise : 289248.500 micros/op 3 ops/sec; 1648.5 MB/s (500000000 per op)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5773
Differential Revision: D17347339
fbshipit-source-id: bfcd74f0f0eb4b322b959be68019ddcaae1e3341
Summary:
Use delete to disable automatic generated methods instead of private, and put the constructor together for more clear.This modification cause the unused field warning, so add unused attribute to disable this warning.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5009
Differential Revision: D17288733
fbshipit-source-id: 8a767ce096f185f1db01bd28fc88fef1cdd921f3
Summary:
Bug found by valgrind. New DynamicBloom wasn't allocating in
block sizes. New assertion added that probes starting in final word
would be in bounds.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5783
Test Plan: ROCKSDB_VALGRIND_RUN=1 DISABLE_JEMALLOC=1 valgrind --leak-check=full ./dynamic_bloom_test
Differential Revision: D17270623
Pulled By: pdillinger
fbshipit-source-id: 1e0407504b875133a771383cd488c70f91be2b87
Summary:
Check that we don't accidentally change the on-disk format of
existing Bloom filter implementations, including for various
CACHE_LINE_SIZE (by changing temporarily).
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5778
Test Plan: thisisthetest
Differential Revision: D17269630
Pulled By: pdillinger
fbshipit-source-id: c77017662f010a77603b7d475892b1f0d5563d8b
Summary:
FullFilterBitsBuilder::CalculateSpace use CACHE_LINE_SIZE which is 64@X86 but 128@ARM64
when it run bloom_test.FullVaryingLengths it failed on ARM64 server,
the assert can be fixed by change 128->CACHE_LINE_SIZE*2 as merged
ASSERT_LE(FilterSize(), (size_t)((length * 10 / 8) + CACHE_LINE_SIZE * 2 + 5)) << length;
run bloom_test
before fix:
/root/rocksdb-master/util/bloom_test.cc:281: Failure
Expected: (FilterSize()) <= ((size_t)((length * 10 / 8) + 128 + 5)), actual: 389 vs 383
200
[ FAILED ] FullBloomTest.FullVaryingLengths (32 ms)
[----------] 4 tests from FullBloomTest (32 ms total)
[----------] Global test environment tear-down
[==========] 7 tests from 2 test cases ran. (116 ms total)
[ PASSED ] 6 tests.
[ FAILED ] 1 test, listed below:
[ FAILED ] FullBloomTest.FullVaryingLengths
after fix:
Filters: 37 good, 0 mediocre
[ OK ] FullBloomTest.FullVaryingLengths (90 ms)
[----------] 4 tests from FullBloomTest (90 ms total)
[----------] Global test environment tear-down
[==========] 7 tests from 2 test cases ran. (174 ms total)
[ PASSED ] 7 tests.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5745
Differential Revision: D17076047
fbshipit-source-id: e7beb5d55d4855fceb2b84bc8119a6b0759de635
Summary:
Since DynamicBloom is now only used in-memory, we're free to
change it without schema compatibility issues. The new implementation
is drawn from (with manifest permission)
303542a767/bloom_simulation_tests/foo.cc (L613)
This has several speed advantages over the prior implementation:
* Uses fastrange instead of %
* Minimum logic to determine first (and all) probed memory addresses
* (Major) Two probes per 64-bit memory fetch/write.
* Very fast and effective (murmur-like) hash expansion/re-mixing. (At
least on recent CPUs, integer multiplication is very cheap.)
While a Bloom filter with 512-bit cache locality has about a 1.15x FP
rate penalty (e.g. 0.84% to 0.97%), further restricting to two probes
per 64 bits incurs an additional 1.12x FP rate penalty (e.g. 0.97% to
1.09%). Nevertheless, the unit tests show no "mediocre" FP rate samples,
unlike the old implementation with more erratic FP rates.
Especially for the memtable, we expect speed to outweigh somewhat higher
FP rates. For example, a negative table query would have to be 1000x
slower than a BF query to justify doubling BF query time to shave 10% off
FP rate (working assumption around 1% FP rate). While that seems likely
for SSTs, my data suggests a speed factor of roughly 50x for the memtable
(vs. BF; ~1.5% lower write throughput when enabling memtable Bloom
filter, after this change). Thus, it's probably not worth even 5% more
time in the Bloom filter to shave off 1/10th of the Bloom FP rate, or 0.1%
in absolute terms, and it's probably at least 20% slower to recoup that
much FP rate from this new implementation. Because of this, we do not see
a need for a 'locality' option that affects the MemTable Bloom filter
and have decoupled the MemTable Bloom filter from Options::bloom_locality.
Note that just 3% more memory to the Bloom filter (10.3 bits per key vs.
just 10) is able to make up for the ~12% FP rate drop in the new
implementation:
[] # Nearly "ideal" FP-wise but reasonably fast cache-local implementation
[~/wormhashing/bloom_simulation_tests] ./foo_gcc_IMPL_CACHE_WORM64_FROM32_any.out 10000000 6 10 $RANDOM 100000000
./foo_gcc_IMPL_CACHE_WORM64_FROM32_any.out time: 3.29372 sampled_fp_rate: 0.00985956 ...
[] # Close match to this new implementation
[~/wormhashing/bloom_simulation_tests] ./foo_gcc_IMPL_CACHE_MUL64_BLOCK_FROM32_any.out 10000000 6 10.3 $RANDOM 100000000
./foo_gcc_IMPL_CACHE_MUL64_BLOCK_FROM32_any.out time: 2.10072 sampled_fp_rate: 0.00985655 ...
[] # Old locality=1 implementation
[~/wormhashing/bloom_simulation_tests] ./foo_gcc_IMPL_CACHE_ROCKSDB_DYNAMIC_any.out 10000000 6 10 $RANDOM 100000000
./foo_gcc_IMPL_CACHE_ROCKSDB_DYNAMIC_any.out time: 3.95472 sampled_fp_rate: 0.00988943 ...
Also note the dramatic speed improvement vs. alternatives.
--
Performance unit test: DynamicBloomTest.concurrent_with_perf is updated
to report more precise timing data. (Measure running time of each
thread, not just longest running thread, etc.) Results averaged over
various sizes enabled with --enable_perf and 20 runs each; old dynamic
bloom refers to locality=1, the faster of the old:
old dynamic bloom, avg add latency = 65.6468
new dynamic bloom, avg add latency = 44.3809
old dynamic bloom, avg query latency = 50.6485
new dynamic bloom, avg query latency = 43.2186
old avg parallel add latency = 41.678
new avg parallel add latency = 24.5238
old avg parallel hit latency = 14.6322
new avg parallel hit latency = 12.3939
old avg parallel miss latency = 16.7289
new avg parallel miss latency = 12.2134
Tested on a dedicated 64-bit production machine at Facebook. Significant
improvement all around.
Despite now using std::atomic<uint64_t>, quick before-and-after test on
a 32-bit machine (Intel Atom N270, released 2008) shows no regression in
performance, in some cases modest improvement.
--
Performance integration test (synthetic): with DEBUG_LEVEL=0, used
TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=fillrandom,readmissing,readrandom,stats --num=2000000
and optionally with -memtable_whole_key_filtering -memtable_bloom_size_ratio=0.01
300 runs each configuration.
Write throughput change by enabling memtable bloom:
Old locality=0: -3.06%
Old locality=1: -2.37%
New: -1.50%
conclusion -> seems to substantially close the gap
Readmissing throughput change by enabling memtable bloom:
Old locality=0: +34.47%
Old locality=1: +34.80%
New: +33.25%
conclusion -> maybe a small new penalty from FP rate
Readrandom throughput change by enabling memtable bloom:
Old locality=0: +31.54%
Old locality=1: +31.13%
New: +30.60%
conclusion -> maybe also from FP rate (after memtable flush)
--
Another conclusion we can draw from this new implementation is that the
existing 32-bit hash function is not inherently crippling the Bloom
filter speed or accuracy, below about 5 million keys. For speed, the
implementation is essentially the same whether starting with 32-bits or
64-bits of hash; it just determines whether the first multiplication
after fastrange is a pseudorandom expansion or needed re-mix. Note that
this multiplication can occur while memory is fetching.
For accuracy, in a standard configuration, you need about 5 million
keys before you have about a 1.1x FP penalty due to using a
32-bit hash vs. 64-bit:
[~/wormhashing/bloom_simulation_tests] ./foo_gcc_IMPL_CACHE_MUL64_BLOCK_FROM32_any.out $((5 * 1000 * 1000 * 10)) 6 10 $RANDOM 100000000
./foo_gcc_IMPL_CACHE_MUL64_BLOCK_FROM32_any.out time: 2.52069 sampled_fp_rate: 0.0118267 ...
[~/wormhashing/bloom_simulation_tests] ./foo_gcc_IMPL_CACHE_MUL64_BLOCK_any.out $((5 * 1000 * 1000 * 10)) 6 10 $RANDOM 100000000
./foo_gcc_IMPL_CACHE_MUL64_BLOCK_any.out time: 2.43871 sampled_fp_rate: 0.0109059
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5762
Differential Revision: D17214194
Pulled By: pdillinger
fbshipit-source-id: ad9da031772e985fd6b62a0e1db8e81892520595
Summary:
DynamicBloom was being used both for memory-only and for on-disk filters, as part of the PlainTable format. To set up enhancements to the memtable Bloom filter, this splits the code into two copies and removes unused features from each copy. Adds test PlainTableDBTest.BloomSchema to ensure no accidental change to that format.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5767
Differential Revision: D17206963
Pulled By: pdillinger
fbshipit-source-id: 6cce8d55305ed0df051b4c58bdc98c8ad81d0553
Summary:
MyRocks currently sets `max_write_buffer_number_to_maintain` in order to maintain enough history for transaction conflict checking. The effectiveness of this approach depends on the size of memtables. When memtables are small, it may not keep enough history; when memtables are large, this may consume too much memory.
We are proposing a new way to configure memtable list history: by limiting the memory usage of immutable memtables. The new option is `max_write_buffer_size_to_maintain` and it will take precedence over the old `max_write_buffer_number_to_maintain` if they are both set to non-zero values. The new option accounts for the total memory usage of flushed immutable memtables and mutable memtable. When the total usage exceeds the limit, RocksDB may start dropping immutable memtables (which is also called trimming history), starting from the oldest one.
The semantics of the old option actually works both as an upper bound and lower bound. History trimming will start if number of immutable memtables exceeds the limit, but it will never go below (limit-1) due to history trimming.
In order the mimic the behavior with the new option, history trimming will stop if dropping the next immutable memtable causes the total memory usage go below the size limit. For example, assuming the size limit is set to 64MB, and there are 3 immutable memtables with sizes of 20, 30, 30. Although the total memory usage is 80MB > 64MB, dropping the oldest memtable will reduce the memory usage to 60MB < 64MB, so in this case no memtable will be dropped.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5022
Differential Revision: D14394062
Pulled By: miasantreble
fbshipit-source-id: 60457a509c6af89d0993f988c9b5c2aa9e45f5c5
Peter Dillinger