rocksdb/table/block_based
Peter Dillinger e4f1e64c30 Add prefetching (batched MultiGet) for experimental Ribbon filter (#7889)
Summary:
Adds support for prefetching data in Ribbon queries,
which especially optimizes batched Ribbon queries for MultiGet
(~222ns/key to ~97ns/key) but also single key queries on cold memory
(~333ns to ~226ns) because many queries span more than one cache line.

This required some refactoring of the query algorithm, and there
does not appear to be a noticeable regression in "hot memory" query
times (perhaps from 48ns to 50ns).

Pull Request resolved: https://github.com/facebook/rocksdb/pull/7889

Test Plan:
existing unit tests, plus performance validation with
filter_bench:

Each data point is the best of two runs. I saturated the machine
CPUs with other filter_bench runs in the background.

Before:

    $ ./filter_bench -impl=3 -m_keys_total_max=200 -average_keys_per_filter=100000 -m_queries=50
    WARNING: Assertions are enabled; benchmarks unnecessarily slow
    Building...
    Build avg ns/key: 125.86
    Number of filters: 1993
    Total size (MB): 168.166
    Reported total allocated memory (MB): 183.211
    Reported internal fragmentation: 8.94626%
    Bits/key stored: 7.05341
    Prelim FP rate %: 0.951827
    ----------------------------
    Mixed inside/outside queries...
      Single filter net ns/op: 48.0111
      Batched, prepared net ns/op: 222.384
      Batched, unprepared net ns/op: 343.908
      Skewed 50% in 1% net ns/op: 252.916
      Skewed 80% in 20% net ns/op: 320.579
      Random filter net ns/op: 332.957

After:

    $ ./filter_bench -impl=3 -m_keys_total_max=200 -average_keys_per_filter=100000 -m_queries=50
    WARNING: Assertions are enabled; benchmarks unnecessarily slow
    Building...
    Build avg ns/key: 128.117
    Number of filters: 1993
    Total size (MB): 168.166
    Reported total allocated memory (MB): 183.211
    Reported internal fragmentation: 8.94626%
    Bits/key stored: 7.05341
    Prelim FP rate %: 0.951827
    ----------------------------
    Mixed inside/outside queries...
      Single filter net ns/op: 49.8812
      Batched, prepared net ns/op: 97.1514
      Batched, unprepared net ns/op: 222.025
      Skewed 50% in 1% net ns/op: 197.48
      Skewed 80% in 20% net ns/op: 212.457
      Random filter net ns/op: 226.464

Bloom comparison, for reference:

    $ ./filter_bench -impl=2 -m_keys_total_max=200 -average_keys_per_filter=100000 -m_queries=50
    WARNING: Assertions are enabled; benchmarks unnecessarily slow
    Building...
    Build avg ns/key: 35.3042
    Number of filters: 1993
    Total size (MB): 238.488
    Reported total allocated memory (MB): 262.875
    Reported internal fragmentation: 10.2255%
    Bits/key stored: 10.0029
    Prelim FP rate %: 0.965327
    ----------------------------
    Mixed inside/outside queries...
      Single filter net ns/op: 9.09931
      Batched, prepared net ns/op: 34.21
      Batched, unprepared net ns/op: 88.8564
      Skewed 50% in 1% net ns/op: 139.75
      Skewed 80% in 20% net ns/op: 181.264
      Random filter net ns/op: 173.88

Reviewed By: jay-zhuang

Differential Revision: D26378710

Pulled By: pdillinger

fbshipit-source-id: 058428967c55ed763698284cd3b4bbe3351b6e69
2021-02-10 21:04:56 -08:00
..
binary_search_index_reader.cc Separate internal and user key comparators in BlockIter (#6944) 2020-07-07 17:26:16 -07:00
binary_search_index_reader.h Extend Get/MultiGet deadline support to table open (#6982) 2020-06-29 14:53:17 -07:00
block.cc Remove unused includes (#7604) 2020-10-28 23:22:27 -07:00
block.h Add EnvTestWithParam::OptionsTest to the ASSERT_STATUS_CHECKED passes (#7283) 2020-08-20 19:18:35 -07:00
block_based_filter_block.cc Exclude timestamp from prefix extractor (#7668) 2020-12-01 14:07:15 -08:00
block_based_filter_block.h Exclude timestamp from prefix extractor (#7668) 2020-12-01 14:07:15 -08:00
block_based_filter_block_test.cc
block_based_table_builder.cc Add a SystemClock class to capture the time functions of an Env (#7858) 2021-01-25 22:09:11 -08:00
block_based_table_builder.h Make parallel compression optimization code tidier (#6888) 2020-10-22 11:05:25 -07:00
block_based_table_factory.cc Fix the logic of setting read_amp_bytes_per_bit from OPTIONS file (#7680) 2020-11-17 00:44:30 -08:00
block_based_table_factory.h Create a Customizable class to load classes and configurations (#6590) 2020-11-11 15:10:41 -08:00
block_based_table_iterator.cc Clean up InternalIterator upper bound logic a little bit (#7200) 2020-08-05 10:44:57 -07:00
block_based_table_iterator.h Exclude timestamp from prefix extractor (#7668) 2020-12-01 14:07:15 -08:00
block_based_table_reader.cc Add a SystemClock class to capture the time functions of an Env (#7858) 2021-01-25 22:09:11 -08:00
block_based_table_reader.h Add a SystemClock class to capture the time functions of an Env (#7858) 2021-01-25 22:09:11 -08:00
block_based_table_reader_impl.h Divide block_based_table_reader.cc (#6527) 2020-03-12 21:41:50 -07:00
block_based_table_reader_test.cc Add a SystemClock class to capture the time functions of an Env (#7858) 2021-01-25 22:09:11 -08:00
block_builder.cc Add pipelined & parallel compression optimization (#6262) 2020-04-01 16:40:18 -07:00
block_builder.h Add pipelined & parallel compression optimization (#6262) 2020-04-01 16:40:18 -07:00
block_prefetcher.cc Add buffer prefetch support for non directIO usecase (#7312) 2020-08-27 18:16:53 -07:00
block_prefetcher.h De-template block based table iterator (#6531) 2020-03-16 12:20:50 -07:00
block_prefix_index.cc
block_prefix_index.h
block_test.cc More Makefile Cleanup (#7097) 2020-07-09 14:35:17 -07:00
block_type.h
cachable_entry.h
data_block_footer.cc
data_block_footer.h
data_block_hash_index.cc
data_block_hash_index.h
data_block_hash_index_test.cc Make StringEnv, StringSink, StringSource use FS classes (#7786) 2021-01-04 16:01:01 -08:00
filter_block.h Exclude timestamp from prefix extractor (#7668) 2020-12-01 14:07:15 -08:00
filter_block_reader_common.cc
filter_block_reader_common.h
filter_policy.cc Add prefetching (batched MultiGet) for experimental Ribbon filter (#7889) 2021-02-10 21:04:56 -08:00
filter_policy_internal.h Support optimize_filters_for_memory for Ribbon filter (#7774) 2020-12-18 14:31:03 -08:00
flush_block_policy.cc
flush_block_policy.h
full_filter_block.cc Exclude timestamp from prefix extractor (#7668) 2020-12-01 14:07:15 -08:00
full_filter_block.h Exclude timestamp from prefix extractor (#7668) 2020-12-01 14:07:15 -08:00
full_filter_block_test.cc Use size_t for filter APIs, protect against overflow (#7726) 2020-12-11 22:18:12 -08:00
hash_index_reader.cc Separate internal and user key comparators in BlockIter (#6944) 2020-07-07 17:26:16 -07:00
hash_index_reader.h Extend Get/MultiGet deadline support to table open (#6982) 2020-06-29 14:53:17 -07:00
index_builder.cc Move break into block (#7468) 2020-09-30 20:24:23 -07:00
index_builder.h Make db_basic_test pass assert status checked (#7452) 2020-09-29 09:49:04 -07:00
index_reader_common.cc Divide block_based_table_reader.cc (#6527) 2020-03-12 21:41:50 -07:00
index_reader_common.h Divide block_based_table_reader.cc (#6527) 2020-03-12 21:41:50 -07:00
mock_block_based_table.h For ApproximateSizes, pro-rate table metadata size over data blocks (#6784) 2020-06-02 12:30:23 -07:00
parsed_full_filter_block.cc
parsed_full_filter_block.h
partitioned_filter_block.cc Add a SystemClock class to capture the time functions of an Env (#7858) 2021-01-25 22:09:11 -08:00
partitioned_filter_block.h Return error if Get/Multi() fails in Prefetching Filter blocks (#7543) 2020-10-14 10:45:36 -07:00
partitioned_filter_block_test.cc Remove unused includes (#7604) 2020-10-28 23:22:27 -07:00
partitioned_index_iterator.cc Fix misspelling of PartitionedIndexIterator (#7450) 2020-09-29 16:28:13 -07:00
partitioned_index_iterator.h Fix misspelling of PartitionedIndexIterator (#7450) 2020-09-29 16:28:13 -07:00
partitioned_index_reader.cc Add a SystemClock class to capture the time functions of an Env (#7858) 2021-01-25 22:09:11 -08:00
partitioned_index_reader.h Get() to fail with underlying failures in PartitionIndexReader::CacheDependencies() (#7297) 2020-08-25 19:01:05 -07:00
reader_common.cc Fix block checksum for >=4GB, refactor (#6978) 2020-06-19 16:18:24 -07:00
reader_common.h Bring the Configurable options together (#5753) 2020-09-14 17:01:01 -07:00
uncompression_dict_reader.cc Extend Get/MultiGet deadline support to table open (#6982) 2020-06-29 14:53:17 -07:00
uncompression_dict_reader.h Extend Get/MultiGet deadline support to table open (#6982) 2020-06-29 14:53:17 -07:00