rocksdb/db
sdong b397dcd390 Change The Way Level Target And Compaction Score Are Calculated (#10057)
Summary:
The current level targets for dynamical leveling has a problem: the target level size will dramatically change after a L0->L1 compaction. When there are many L0 bytes, lower level compactions are delayed, but they will be resumed after the L0->L1 compaction finishes, so the expected write amplification benefits might not be realized. The proposal here is to revert the level targetting size, but instead relying on adjusting score for each level to prioritize levels that need to compact most.
Basic idea:
(1) target level size isn't adjusted, but score is adjusted. The reasoning is that with parallel compactions, holding compactions from happening might not be desirable, but we would like the compactions are scheduled from the level we feel most needed. For example, if we have a extra-large L2, we would like all compactions are scheduled for L2->L3 compactions, rather than L4->L5. This gets complicated when a large L0->L1 compaction is going on. Should we compact L2->L3 or L4->L5. So the proposal for that is:
(2) the score is calculated by actual level size / (target size + estimated upper bytes coming down). The reasoning is that if we have a large amount of pending L0/L1 bytes coming down, compacting L2->L3 might be more expensive, as when the L0 bytes are compacted down to L2, the actual L2->L3 fanout would change dramatically. On the other hand, when the amount of bytes coming down to L5, the impacts to L5->L6 fanout are much less. So when calculating target score, we can adjust it by adding estimated downward bytes to the target level size.

Pull Request resolved: https://github.com/facebook/rocksdb/pull/10057

Test Plan: Repurpose tests VersionStorageInfoTest.MaxBytesForLevelDynamicWithLargeL0_* tests to cover this scenario.

Reviewed By: ajkr

Differential Revision: D37539742

fbshipit-source-id: 9c154cbfe92023f918cf5d80875d8776ad4831a4
2022-06-30 13:32:47 -07:00
..
blob Enable blob caching for MultiGetBlob in RocksDB (#10272) 2022-06-30 13:24:35 -07:00
compaction Add API for writing wide-column entities (#10242) 2022-06-25 15:30:47 -07:00
db_impl Add API for writing wide-column entities (#10242) 2022-06-25 15:30:47 -07:00
wide Add API for writing wide-column entities (#10242) 2022-06-25 15:30:47 -07:00
arena_wrapped_db_iter.cc Remove own ToString() (#9955) 2022-05-06 13:03:58 -07:00
arena_wrapped_db_iter.h Cleanup includes in dbformat.h (#8930) 2021-09-29 04:04:40 -07:00
builder.cc Consolidate manual_compaction_paused_ check (#10070) 2022-06-06 18:32:26 -07:00
builder.h CompactionIterator sees consistent view of which keys are committed (#9830) 2022-04-14 11:11:04 -07:00
c.cc Add load_latest_options() to C api (#10152) 2022-06-30 11:03:52 -07:00
c_test.c Add load_latest_options() to C api (#10152) 2022-06-30 11:03:52 -07:00
column_family.cc Fix A Bug Where Concurrent Compactions Cause Further Slowing Down (#10270) 2022-06-29 11:20:36 -07:00
column_family.h Dynamically changeable `MemPurge` option (#10011) 2022-06-23 09:42:18 -07:00
column_family_test.cc Deflake column_family_test to avoid hang (#10060) 2022-05-30 12:31:46 -07:00
compact_files_test.cc Remove own ToString() (#9955) 2022-05-06 13:03:58 -07:00
comparator_db_test.cc Document design/specification bugs with auto_prefix_mode (#10144) 2022-06-13 11:08:50 -07:00
convenience.cc Specify largest_seqno in VerifyChecksum (#9919) 2022-05-02 10:22:08 -07:00
corruption_test.cc Persist the new MANIFEST after successfully syncing the new WAL during recovery (#9922) 2022-06-01 10:52:26 -07:00
cuckoo_table_db_test.cc Remove own ToString() (#9955) 2022-05-06 13:03:58 -07:00
db_basic_test.cc Don't count no prefix as Bloom hit (#10244) 2022-06-23 11:00:27 -07:00
db_block_cache_test.cc Clock cache (#10273) 2022-06-29 21:50:39 -07:00
db_bloom_filter_test.cc Update/clarify required properties for prefix extractors (#10245) 2022-06-28 16:08:30 -07:00
db_compaction_filter_test.cc Remove own ToString() (#9955) 2022-05-06 13:03:58 -07:00
db_compaction_test.cc Fix the flaky cursor persist test (#10250) 2022-06-24 14:02:33 -07:00
db_dynamic_level_test.cc Remove deprecated API AdvancedColumnFamilyOptions::soft_rate_limit/hard_rate_limit (#9452) 2022-01-27 13:01:09 -08:00
db_encryption_test.cc Fix a minor issue with initializing the test path (#8555) 2021-07-23 08:38:45 -07:00
db_filesnapshot.cc Reduce risk of backup or checkpoint missing a WAL file (#10083) 2022-06-01 11:02:27 -07:00
db_flush_test.cc Dynamically changeable `MemPurge` option (#10011) 2022-06-23 09:42:18 -07:00
db_info_dumper.cc Printing IO Error in DumpDBFileSummary (#9940) 2022-05-04 10:19:53 -07:00
db_info_dumper.h Add a DB Session ID (#6959) 2020-06-15 10:47:02 -07:00
db_inplace_update_test.cc Fix in-place updates for value types other than kTypeValue (#10254) 2022-06-27 16:37:09 -07:00
db_io_failure_test.cc Enable a few unit tests to use custom Env objects (#9087) 2021-11-08 11:05:59 -08:00
db_iter.cc Fix assertion error with read_opts.iter_start_ts (#10279) 2022-06-30 10:16:03 -07:00
db_iter.h Support `iter_start_ts` for backward iteration (#10200) 2022-06-28 19:51:05 -07:00
db_iter_stress_test.cc Remove own ToString() (#9955) 2022-05-06 13:03:58 -07:00
db_iter_test.cc Remove own ToString() (#9955) 2022-05-06 13:03:58 -07:00
db_iterator_test.cc Remove own ToString() (#9955) 2022-05-06 13:03:58 -07:00
db_kv_checksum_test.cc Add API for writing wide-column entities (#10242) 2022-06-25 15:30:47 -07:00
db_log_iter_test.cc Remove own ToString() (#9955) 2022-05-06 13:03:58 -07:00
db_logical_block_size_cache_test.cc Attempt to deflake DBLogicalBlockSizeCacheTest.CreateColumnFamilies (#9516) 2022-03-04 11:35:28 -08:00
db_memtable_test.cc Remove own ToString() (#9955) 2022-05-06 13:03:58 -07:00
db_merge_operand_test.cc Fix GetMergeOperands() heap-use-after-free on flushed memtable (#9805) 2022-04-05 12:26:36 -07:00
db_merge_operator_test.cc Fix a minor issue with initializing the test path (#8555) 2021-07-23 08:38:45 -07:00
db_options_test.cc Fix typo in comments and code (#10233) 2022-06-22 15:45:21 -07:00
db_properties_test.cc Add blob cache tickers, perf context statistics, and DB properties (#10203) 2022-06-28 13:52:35 -07:00
db_range_del_test.cc Remove own ToString() (#9955) 2022-05-06 13:03:58 -07:00
db_rate_limiter_test.cc Add rate-limiting support to batched MultiGet() (#10159) 2022-06-17 16:40:47 -07:00
db_readonly_with_timestamp_test.cc Add timestamp support to CompactedDBImpl (#10030) 2022-05-24 12:14:10 -07:00
db_secondary_test.cc Fail DB::Open() if logger cannot be created (#9984) 2022-05-27 07:23:31 -07:00
db_sst_test.cc Pass the size of blob files to SstFileManager during DB open (#10062) 2022-05-27 05:58:43 -07:00
db_statistics_test.cc Bytes read stat for `VerifyChecksum()` and `VerifyFileChecksums()` APIs (#8741) 2021-09-07 13:28:29 -07:00
db_table_properties_test.cc Remove own ToString() (#9955) 2022-05-06 13:03:58 -07:00
db_tailing_iter_test.cc Fix a minor issue with initializing the test path (#8555) 2021-07-23 08:38:45 -07:00
db_test.cc Add API for writing wide-column entities (#10242) 2022-06-25 15:30:47 -07:00
db_test2.cc Document design/specification bugs with auto_prefix_mode (#10144) 2022-06-13 11:08:50 -07:00
db_test_util.cc Account memory of FileMetaData in global memory limit (#9924) 2022-06-14 13:06:40 -07:00
db_test_util.h Explicitly closing all directory file descriptors (#10049) 2022-06-01 18:03:34 -07:00
db_universal_compaction_test.cc Remove own ToString() (#9955) 2022-05-06 13:03:58 -07:00
db_wal_test.cc Destroy iniital db dir for a test in DBWALTest (#10221) 2022-06-21 11:27:10 -07:00
db_with_timestamp_basic_test.cc Fix assertion error with read_opts.iter_start_ts (#10279) 2022-06-30 10:16:03 -07:00
db_with_timestamp_compaction_test.cc Use the comparator from the sst file table properties in sst_dump_tool (#9491) 2022-02-08 12:15:35 -08:00
db_with_timestamp_test_util.cc Add timestamp support to DBImplReadOnly (#10004) 2022-05-19 18:39:41 -07:00
db_with_timestamp_test_util.h Add timestamp support to DBImplReadOnly (#10004) 2022-05-19 18:39:41 -07:00
db_write_buffer_manager_test.cc Enable a few unit tests to use custom Env objects (#9087) 2021-11-08 11:05:59 -08:00
db_write_test.cc Fix race condition with WAL tracking and `FlushWAL(true /* sync */)` (#10185) 2022-06-17 16:45:28 -07:00
dbformat.cc Add new value value type for wide-column entities (#10211) 2022-06-20 18:04:08 -07:00
dbformat.h Add new value value type for wide-column entities (#10211) 2022-06-20 18:04:08 -07:00
dbformat_test.cc Enable a few unit tests to use custom Env objects (#9087) 2021-11-08 11:05:59 -08:00
deletefile_test.cc Remove own ToString() (#9955) 2022-05-06 13:03:58 -07:00
error_handler.cc Expand auto recovery to background read errors (#9679) 2022-03-15 14:45:34 -07:00
error_handler.h Expand auto recovery to background read errors (#9679) 2022-03-15 14:45:34 -07:00
error_handler_fs_test.cc Remove own ToString() (#9955) 2022-05-06 13:03:58 -07:00
event_helpers.cc Fix a race condition in WAL tracking causing DB open failure (#9715) 2022-03-23 19:41:31 -07:00
event_helpers.h Add a listener callback for end of auto error recovery (#9244) 2021-12-08 14:30:57 -08:00
experimental.cc Track SST unique id in MANIFEST and verify (#9990) 2022-05-19 11:04:21 -07:00
external_sst_file_basic_test.cc Fix unittest ExternalSSTFileBasicTest.StableSnapshotWhileLoggingToManifest (#10066) 2022-05-31 08:48:57 -07:00
external_sst_file_ingestion_job.cc Track SST unique id in MANIFEST and verify (#9990) 2022-05-19 11:04:21 -07:00
external_sst_file_ingestion_job.h Track SST unique id in MANIFEST and verify (#9990) 2022-05-19 11:04:21 -07:00
external_sst_file_test.cc Remove own ToString() (#9955) 2022-05-06 13:03:58 -07:00
fault_injection_test.cc Fix a bug causing duplicate trailing entries in WritableFile (buffered IO) (#9236) 2021-12-13 09:00:36 -08:00
file_indexer.cc Replace namespace name "rocksdb" with ROCKSDB_NAMESPACE (#6433) 2020-02-20 12:09:57 -08:00
file_indexer.h Use std::numeric_limits<> (#9954) 2022-05-05 13:08:21 -07:00
file_indexer_test.cc Replace namespace name "rocksdb" with ROCKSDB_NAMESPACE (#6433) 2020-02-20 12:09:57 -08:00
filename_test.cc fixing issue #8345 RocksDB does not work when using UNC network paths (#9384) 2022-03-30 15:55:31 -07:00
flush_job.cc Dynamically changeable `MemPurge` option (#10011) 2022-06-23 09:42:18 -07:00
flush_job.h Dynamically changeable `MemPurge` option (#10011) 2022-06-23 09:42:18 -07:00
flush_job_test.cc Add blob source to retrieve blobs in RocksDB (#10198) 2022-06-20 20:58:11 -07:00
flush_scheduler.cc Replace namespace name "rocksdb" with ROCKSDB_NAMESPACE (#6433) 2020-02-20 12:09:57 -08:00
flush_scheduler.h Include C++ standard library headers instead of C compatibility headers (#8068) 2021-03-19 12:09:47 -07:00
forward_iterator.cc Remove own ToString() (#9955) 2022-05-06 13:03:58 -07:00
forward_iterator.h Fast path for detecting unchanged prefix_extractor (#9407) 2022-01-21 11:37:46 -08:00
forward_iterator_bench.cc Remove using namespace (#9369) 2022-01-12 09:31:12 -08:00
history_trimming_iterator.h Add OpenAndTrimHistory API to support trimming data with specified timestamp (#9410) 2022-03-11 16:13:23 -08:00
import_column_family_job.cc Track SST unique id in MANIFEST and verify (#9990) 2022-05-19 11:04:21 -07:00
import_column_family_job.h New stable, fixed-length cache keys (#9126) 2021-12-16 17:15:13 -08:00
import_column_family_test.cc Track SST unique id in MANIFEST and verify (#9990) 2022-05-19 11:04:21 -07:00
internal_stats.cc Add blob cache tickers, perf context statistics, and DB properties (#10203) 2022-06-28 13:52:35 -07:00
internal_stats.h Add blob cache tickers, perf context statistics, and DB properties (#10203) 2022-06-28 13:52:35 -07:00
job_context.h CompactionIterator sees consistent view of which keys are committed (#9830) 2022-04-14 11:11:04 -07:00
kv_checksum.h fix compile errors in db/kv_checksum.h (#9173) 2021-11-16 10:20:50 -08:00
listener_test.cc Remove own ToString() (#9955) 2022-05-06 13:03:58 -07:00
log_format.h Add record to set WAL compression type if enabled (#9556) 2022-02-17 16:19:31 -08:00
log_reader.cc Support read rate-limiting in SequentialFileReader (#9973) 2022-05-24 10:28:57 -07:00
log_reader.h Fix bug for WalManager with compressed WAL (#10130) 2022-06-08 14:16:43 -07:00
log_test.cc Integrate WAL compression into log reader/writer. (#9642) 2022-03-09 15:49:53 -08:00
log_writer.cc Integrate WAL compression into log reader/writer. (#9642) 2022-03-09 15:49:53 -08:00
log_writer.h Fix typo in comments and code (#10233) 2022-06-22 15:45:21 -07:00
logs_with_prep_tracker.cc Replace namespace name "rocksdb" with ROCKSDB_NAMESPACE (#6433) 2020-02-20 12:09:57 -08:00
logs_with_prep_tracker.h Include C++ standard library headers instead of C compatibility headers (#8068) 2021-03-19 12:09:47 -07:00
lookup_key.h Cleanup includes in dbformat.h (#8930) 2021-09-29 04:04:40 -07:00
malloc_stats.cc Replace most typedef with using= (#8751) 2021-09-07 11:31:59 -07:00
malloc_stats.h Replace namespace name "rocksdb" with ROCKSDB_NAMESPACE (#6433) 2020-02-20 12:09:57 -08:00
manual_compaction_test.cc Remove using namespace (#9369) 2022-01-12 09:31:12 -08:00
memtable.cc Fix in-place updates for value types other than kTypeValue (#10254) 2022-06-27 16:37:09 -07:00
memtable.h Fix in-place updates for value types other than kTypeValue (#10254) 2022-06-27 16:37:09 -07:00
memtable_list.cc Encode min_log_number_to_keep and delete_wals_before in one version edit (#9766) 2022-03-31 20:00:52 -07:00
memtable_list.h Fix various spelling errors still found in code (#9653) 2022-05-05 19:45:32 -07:00
memtable_list_test.cc Add blob source to retrieve blobs in RocksDB (#10198) 2022-06-20 20:58:11 -07:00
merge_context.h Add Merge Operator support to WriteBatchWithIndex (#8135) 2021-05-10 12:50:25 -07:00
merge_helper.cc Add API for writing wide-column entities (#10242) 2022-06-25 15:30:47 -07:00
merge_helper.h Support readahead during compaction for blob files (#9187) 2021-11-19 17:53:47 -08:00
merge_helper_test.cc Support readahead during compaction for blob files (#9187) 2021-11-19 17:53:47 -08:00
merge_operator.cc Replace namespace name "rocksdb" with ROCKSDB_NAMESPACE (#6433) 2020-02-20 12:09:57 -08:00
merge_test.cc Make the Env class Customizable (#9293) 2022-01-04 16:45:49 -08:00
obsolete_files_test.cc Remove own ToString() (#9955) 2022-05-06 13:03:58 -07:00
options_file_test.cc No elide constructors (#7798) 2020-12-23 16:55:53 -08:00
output_validator.cc Cleanup includes in dbformat.h (#8930) 2021-09-29 04:04:40 -07:00
output_validator.h Cleanup includes in dbformat.h (#8930) 2021-09-29 04:04:40 -07:00
perf_context_test.cc Remove own ToString() (#9955) 2022-05-06 13:03:58 -07:00
periodic_work_scheduler.cc Fix a timer crash caused by invalid memory management (#9656) 2022-03-12 11:45:56 -08:00
periodic_work_scheduler.h Fix a timer crash caused by invalid memory management (#9656) 2022-03-12 11:45:56 -08:00
periodic_work_scheduler_test.cc Fix a minor issue with initializing the test path (#8555) 2021-07-23 08:38:45 -07:00
pinned_iterators_manager.h Replace most typedef with using= (#8751) 2021-09-07 11:31:59 -07:00
plain_table_db_test.cc Remove own ToString() (#9955) 2022-05-06 13:03:58 -07:00
post_memtable_callback.h Snapshots with user-specified timestamps (#9879) 2022-06-10 16:07:03 -07:00
pre_release_callback.h Fix and detect headers with missing dependencies (#8893) 2021-09-10 10:00:26 -07:00
prefix_test.cc Remove own ToString() (#9955) 2022-05-06 13:03:58 -07:00
range_del_aggregator.cc In ParseInternalKey(), include corrupt key info in Status (#7515) 2020-10-28 10:12:58 -07:00
range_del_aggregator.h Fix some typos in comments (#8066) 2021-03-25 21:18:08 -07:00
range_del_aggregator_bench.cc Cleanup multiple implementations of VectorIterator (#8901) 2021-10-06 07:48:31 -07:00
range_del_aggregator_test.cc Cleanup multiple implementations of VectorIterator (#8901) 2021-10-06 07:48:31 -07:00
range_tombstone_fragmenter.cc Added memtable garbage statistics (#8411) 2021-06-18 04:57:27 -07:00
range_tombstone_fragmenter.h Added memtable garbage statistics (#8411) 2021-06-18 04:57:27 -07:00
range_tombstone_fragmenter_test.cc Cleanup multiple implementations of VectorIterator (#8901) 2021-10-06 07:48:31 -07:00
read_callback.h Fix and detect headers with missing dependencies (#8893) 2021-09-10 10:00:26 -07:00
repair.cc Add blob source to retrieve blobs in RocksDB (#10198) 2022-06-20 20:58:11 -07:00
repair_test.cc Fix failed VerifySstUniqueIds unittests (#10043) 2022-05-24 09:00:06 -07:00
snapshot_checker.h Use STATIC_AVOID_DESTRUCTION for static objects with non-trivial destructors (#9958) 2022-05-17 09:39:22 -07:00
snapshot_impl.cc Replace namespace name "rocksdb" with ROCKSDB_NAMESPACE (#6433) 2020-02-20 12:09:57 -08:00
snapshot_impl.h Snapshots with user-specified timestamps (#9879) 2022-06-10 16:07:03 -07:00
table_cache.cc Fix crash due to MultiGet async IO and direct IO (#10024) 2022-05-20 12:38:21 -07:00
table_cache.h Multi file concurrency in MultiGet using coroutines and async IO (#9968) 2022-05-19 15:36:27 -07:00
table_cache_sync_and_async.h Multi file concurrency in MultiGet using coroutines and async IO (#9968) 2022-05-19 15:36:27 -07:00
table_properties_collector.cc Apply `sample_for_compression` to all block-based tables (#8105) 2021-03-25 15:00:45 -07:00
table_properties_collector.h Track each SST's timestamp information as user properties (#9093) 2021-11-19 11:37:06 -08:00
table_properties_collector_test.cc Improve / clean up meta block code & integrity (#9163) 2021-11-18 11:43:44 -08:00
transaction_log_impl.cc Support read rate-limiting in SequentialFileReader (#9973) 2022-05-24 10:28:57 -07:00
transaction_log_impl.h Add checks to GetUpdatesSince (#9459) 2022-04-14 17:12:16 -07:00
trim_history_scheduler.cc Replace namespace name "rocksdb" with ROCKSDB_NAMESPACE (#6433) 2020-02-20 12:09:57 -08:00
trim_history_scheduler.h Replace namespace name "rocksdb" with ROCKSDB_NAMESPACE (#6433) 2020-02-20 12:09:57 -08:00
version_builder.cc Add basic kRoundRobin compaction policy (#10107) 2022-06-21 11:56:53 -07:00
version_builder.h Account memory of FileMetaData in global memory limit (#9924) 2022-06-14 13:06:40 -07:00
version_builder_test.cc Track SST unique id in MANIFEST and verify (#9990) 2022-05-19 11:04:21 -07:00
version_edit.cc include compaction cursors in VersionEdit debug string (#10288) 2022-06-30 12:46:45 -07:00
version_edit.h Fix typo in comments and code (#10233) 2022-06-22 15:45:21 -07:00
version_edit_handler.cc Add blob source to retrieve blobs in RocksDB (#10198) 2022-06-20 20:58:11 -07:00
version_edit_handler.h Fixed manifest_dump issues when printing keys and values containing null characters (#8378) 2021-06-10 12:55:20 -07:00
version_edit_test.cc Track SST unique id in MANIFEST and verify (#9990) 2022-05-19 11:04:21 -07:00
version_set.cc Change The Way Level Target And Compaction Score Are Calculated (#10057) 2022-06-30 13:32:47 -07:00
version_set.h Enable blob caching for MultiGetBlob in RocksDB (#10272) 2022-06-30 13:24:35 -07:00
version_set_sync_and_async.h Enable blob caching for MultiGetBlob in RocksDB (#10272) 2022-06-30 13:24:35 -07:00
version_set_test.cc Change The Way Level Target And Compaction Score Are Calculated (#10057) 2022-06-30 13:32:47 -07:00
version_util.h Add blob source to retrieve blobs in RocksDB (#10198) 2022-06-20 20:58:11 -07:00
wal_edit.cc Always track WAL obsoletion (#7759) 2020-12-09 16:02:12 -08:00
wal_edit.h Use std::numeric_limits<> (#9954) 2022-05-05 13:08:21 -07:00
wal_edit_test.cc Always track WAL obsoletion (#7759) 2020-12-09 16:02:12 -08:00
wal_manager.cc Fix bug for WalManager with compressed WAL (#10130) 2022-06-08 14:16:43 -07:00
wal_manager.h Fix bug for WalManager with compressed WAL (#10130) 2022-06-08 14:16:43 -07:00
wal_manager_test.cc Add blob source to retrieve blobs in RocksDB (#10198) 2022-06-20 20:58:11 -07:00
write_batch.cc Fix in-place updates for value types other than kTypeValue (#10254) 2022-06-27 16:37:09 -07:00
write_batch_base.cc Replace namespace name "rocksdb" with ROCKSDB_NAMESPACE (#6433) 2020-02-20 12:09:57 -08:00
write_batch_internal.h Add API for writing wide-column entities (#10242) 2022-06-25 15:30:47 -07:00
write_batch_test.cc Remove own ToString() (#9955) 2022-05-06 13:03:58 -07:00
write_callback.h Replace namespace name "rocksdb" with ROCKSDB_NAMESPACE (#6433) 2020-02-20 12:09:57 -08:00
write_callback_test.cc Add WriteOptions::protection_bytes_per_key (#10037) 2022-06-16 23:10:07 -07:00
write_controller.cc Revamp WriteController (#8064) 2021-03-18 09:47:31 -07:00
write_controller.h Set Write rate limiter priority dynamically and pass it to FS (#9988) 2022-05-18 00:41:41 -07:00
write_controller_test.cc Revamp WriteController (#8064) 2021-03-18 09:47:31 -07:00
write_thread.cc Verify write batch checksum before WAL (#10114) 2022-06-15 13:43:58 -07:00
write_thread.h Snapshots with user-specified timestamps (#9879) 2022-06-10 16:07:03 -07:00