rocksdb/table
Peter Dillinger 311a544c2a Use deleters to label cache entries and collect stats (#8297)
Summary:
This change gathers and publishes statistics about the
kinds of items in block cache. This is especially important for
profiling relative usage of cache by index vs. filter vs. data blocks.
It works by iterating over the cache during periodic stats dump
(InternalStats, stats_dump_period_sec) or on demand when
DB::Get(Map)Property(kBlockCacheEntryStats), except that for
efficiency and sharing among column families, saved data from
the last scan is used when the data is not considered too old.

The new information can be seen in info LOG, for example:

    Block cache LRUCache@0x7fca62229330 capacity: 95.37 MB collections: 8 last_copies: 0 last_secs: 0.00178 secs_since: 0
    Block cache entry stats(count,size,portion): DataBlock(7092,28.24 MB,29.6136%) FilterBlock(215,867.90 KB,0.888728%) FilterMetaBlock(2,5.31 KB,0.00544%) IndexBlock(217,180.11 KB,0.184432%) WriteBuffer(1,256.00 KB,0.262144%) Misc(1,0.00 KB,0%)

And also through DB::GetProperty and GetMapProperty (here using
ldb just for demonstration):

    $ ./ldb --db=/dev/shm/dbbench/ get_property rocksdb.block-cache-entry-stats
    rocksdb.block-cache-entry-stats.bytes.data-block: 0
    rocksdb.block-cache-entry-stats.bytes.deprecated-filter-block: 0
    rocksdb.block-cache-entry-stats.bytes.filter-block: 0
    rocksdb.block-cache-entry-stats.bytes.filter-meta-block: 0
    rocksdb.block-cache-entry-stats.bytes.index-block: 178992
    rocksdb.block-cache-entry-stats.bytes.misc: 0
    rocksdb.block-cache-entry-stats.bytes.other-block: 0
    rocksdb.block-cache-entry-stats.bytes.write-buffer: 0
    rocksdb.block-cache-entry-stats.capacity: 8388608
    rocksdb.block-cache-entry-stats.count.data-block: 0
    rocksdb.block-cache-entry-stats.count.deprecated-filter-block: 0
    rocksdb.block-cache-entry-stats.count.filter-block: 0
    rocksdb.block-cache-entry-stats.count.filter-meta-block: 0
    rocksdb.block-cache-entry-stats.count.index-block: 215
    rocksdb.block-cache-entry-stats.count.misc: 1
    rocksdb.block-cache-entry-stats.count.other-block: 0
    rocksdb.block-cache-entry-stats.count.write-buffer: 0
    rocksdb.block-cache-entry-stats.id: LRUCache@0x7f3636661290
    rocksdb.block-cache-entry-stats.percent.data-block: 0.000000
    rocksdb.block-cache-entry-stats.percent.deprecated-filter-block: 0.000000
    rocksdb.block-cache-entry-stats.percent.filter-block: 0.000000
    rocksdb.block-cache-entry-stats.percent.filter-meta-block: 0.000000
    rocksdb.block-cache-entry-stats.percent.index-block: 2.133751
    rocksdb.block-cache-entry-stats.percent.misc: 0.000000
    rocksdb.block-cache-entry-stats.percent.other-block: 0.000000
    rocksdb.block-cache-entry-stats.percent.write-buffer: 0.000000
    rocksdb.block-cache-entry-stats.secs_for_last_collection: 0.000052
    rocksdb.block-cache-entry-stats.secs_since_last_collection: 0

Solution detail - We need some way to flag what kind of blocks each
entry belongs to, preferably without changing the Cache API.
One of the complications is that Cache is a general interface that could
have other users that don't adhere to whichever convention we decide
on for keys and values. Or we would pay for an extra field in the Handle
that would only be used for this purpose.

This change uses a back-door approach, the deleter, to indicate the
"role" of a Cache entry (in addition to the value type, implicitly).
This has the added benefit of ensuring proper code origin whenever we
recognize a particular role for a cache entry; if the entry came from
some other part of the code, it will use an unrecognized deleter, which
we simply attribute to the "Misc" role.

An internal API makes for simple instantiation and automatic
registration of Cache deleters for a given value type and "role".

Another internal API, CacheEntryStatsCollector, solves the problem of
caching the results of a scan and sharing them, to ensure scans are
neither excessive nor redundant so as not to harm Cache performance.

Because code is added to BlocklikeTraits, it is pulled out of
block_based_table_reader.cc into its own file.

This is a reformulation of https://github.com/facebook/rocksdb/issues/8276, without the type checking option
(could still be added), and with actual stat gathering.

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8297

Test Plan: manual testing with db_bench, and a couple of basic unit tests

Reviewed By: ltamasi

Differential Revision: D28488721

Pulled By: pdillinger

fbshipit-source-id: 472f524a9691b5afb107934be2d41d84f2b129fb
2021-05-19 16:51:13 -07:00
..
adaptive Refactor: use TableBuilderOptions to reduce parameter lists (#8240) 2021-04-29 07:00:50 -07:00
block_based Use deleters to label cache entries and collect stats (#8297) 2021-05-19 16:51:13 -07:00
cuckoo Make ImmutableOptions struct that inherits from ImmutableCFOptions and ImmutableDBOptions (#8262) 2021-05-05 14:00:17 -07:00
plain Make it possible to apply only a subrange of table property collectors (#8298) 2021-05-17 18:28:39 -07:00
block_fetcher_test.cc Make it possible to apply only a subrange of table property collectors (#8298) 2021-05-17 18:28:39 -07:00
block_fetcher.cc Rename variables in ImmutableCFOptions to avoid conflicts with ImmutableDBOptions (#8227) 2021-04-26 12:43:45 -07:00
block_fetcher.h Make ImmutableOptions struct that inherits from ImmutableCFOptions and ImmutableDBOptions (#8262) 2021-05-05 14:00:17 -07:00
cleanable_test.cc Replace namespace name "rocksdb" with ROCKSDB_NAMESPACE (#6433) 2020-02-20 12:09:57 -08:00
format.cc Make ImmutableOptions struct that inherits from ImmutableCFOptions and ImmutableDBOptions (#8262) 2021-05-05 14:00:17 -07:00
format.h Make ImmutableOptions struct that inherits from ImmutableCFOptions and ImmutableDBOptions (#8262) 2021-05-05 14:00:17 -07:00
get_context.cc Use SystemClock* instead of std::shared_ptr<SystemClock> in lower level routines (#8033) 2021-03-15 04:34:11 -07:00
get_context.h Use SystemClock* instead of std::shared_ptr<SystemClock> in lower level routines (#8033) 2021-03-15 04:34:11 -07:00
internal_iterator.h Clean up InternalIterator upper bound logic a little bit (#7200) 2020-08-05 10:44:57 -07:00
iter_heap.h Replace namespace name "rocksdb" with ROCKSDB_NAMESPACE (#6433) 2020-02-20 12:09:57 -08:00
iterator_wrapper.h Redesign block cache pinning API (#7520) 2020-10-11 14:58:24 -07:00
iterator.cc Replace namespace name "rocksdb" with ROCKSDB_NAMESPACE (#6433) 2020-02-20 12:09:57 -08:00
merger_test.cc More Makefile Cleanup (#7097) 2020-07-09 14:35:17 -07:00
merging_iterator.cc Add some simulator cache and block tracer tests to ASSERT_STATUS_CHECKED (#7305) 2020-08-24 16:43:31 -07:00
merging_iterator.h Replace namespace name "rocksdb" with ROCKSDB_NAMESPACE (#6433) 2020-02-20 12:09:57 -08:00
meta_blocks.cc Make ImmutableOptions struct that inherits from ImmutableCFOptions and ImmutableDBOptions (#8262) 2021-05-05 14:00:17 -07:00
meta_blocks.h Make ImmutableOptions struct that inherits from ImmutableCFOptions and ImmutableDBOptions (#8262) 2021-05-05 14:00:17 -07:00
mock_table.cc Refactor: use TableBuilderOptions to reduce parameter lists (#8240) 2021-04-29 07:00:50 -07:00
mock_table.h Refactor: use TableBuilderOptions to reduce parameter lists (#8240) 2021-04-29 07:00:50 -07:00
multiget_context.h Add initial blob support to batched MultiGet (#7766) 2020-12-14 13:48:22 -08:00
persistent_cache_helper.cc Add more tests to ASSERT_STATUS_CHECKED (#7367) 2020-09-16 15:48:07 -07:00
persistent_cache_helper.h Replace namespace name "rocksdb" with ROCKSDB_NAMESPACE (#6433) 2020-02-20 12:09:57 -08:00
persistent_cache_options.h Replace namespace name "rocksdb" with ROCKSDB_NAMESPACE (#6433) 2020-02-20 12:09:57 -08:00
scoped_arena_iterator.h Replace namespace name "rocksdb" with ROCKSDB_NAMESPACE (#6433) 2020-02-20 12:09:57 -08:00
sst_file_dumper.cc Make it possible to apply only a subrange of table property collectors (#8298) 2021-05-17 18:28:39 -07:00
sst_file_dumper.h Make ImmutableOptions struct that inherits from ImmutableCFOptions and ImmutableDBOptions (#8262) 2021-05-05 14:00:17 -07:00
sst_file_reader_test.cc Add a host location property to TableProperties (#7479) 2020-10-19 11:38:48 -07:00
sst_file_reader.cc Make ImmutableOptions struct that inherits from ImmutableCFOptions and ImmutableDBOptions (#8262) 2021-05-05 14:00:17 -07:00
sst_file_writer_collectors.h Apply sample_for_compression to all block-based tables (#8105) 2021-03-25 15:00:45 -07:00
sst_file_writer.cc Make it possible to apply only a subrange of table property collectors (#8298) 2021-05-17 18:28:39 -07:00
table_builder.h Make it possible to apply only a subrange of table property collectors (#8298) 2021-05-17 18:28:39 -07:00
table_factory.cc Create a Customizable class to load classes and configurations (#6590) 2020-11-11 15:10:41 -08:00
table_properties_internal.h Replace namespace name "rocksdb" with ROCKSDB_NAMESPACE (#6433) 2020-02-20 12:09:57 -08:00
table_properties.cc Add sample_for_compression results to table properties (#8139) 2021-03-31 18:21:50 -07:00
table_reader_bench.cc Make it possible to apply only a subrange of table property collectors (#8298) 2021-05-17 18:28:39 -07:00
table_reader_caller.h Replace namespace name "rocksdb" with ROCKSDB_NAMESPACE (#6433) 2020-02-20 12:09:57 -08:00
table_reader.h dedup ReadOptions in iterator hierarchy (#7210) 2020-08-03 15:23:04 -07:00
table_test.cc Make it possible to apply only a subrange of table property collectors (#8298) 2021-05-17 18:28:39 -07:00
two_level_iterator.cc Redesign block cache pinning API (#7520) 2020-10-11 14:58:24 -07:00
two_level_iterator.h Replace namespace name "rocksdb" with ROCKSDB_NAMESPACE (#6433) 2020-02-20 12:09:57 -08:00