rocksdb/java/rocksjni
Changyu Bi d5345a8ff7 Introduce a transaction option to skip memtable write during commit (#13144)
Summary:
add a new transaction option `TransactionOptions::commit_bypass_memtable` that will ingest the transaction into a DB as an immutable memtables, skipping memtable writes during transaction commit. This helps to reduce the blocking time of committing a large transaction, which is mostly spent on memtable writes. The ingestion is done by creating WBWIMemTable using transaction's underlying WBWI, and ingest it as the latest immutable memtable. The feature will be experimental.

Major changes are:
1. write path change to ingest the transaction, mostly in WriteImpl() and IngestWBWI() in db_impl_write.cc.
2. WBWI changes to track some per CF stats like entry count and overwritten single deletion count, and track which keys have overwritten single deletions (see 3.). Per CF stat is used to precompute the number of entries in each WBWIMemTable.
3. WBWIMemTable Iterator changes to emit overwritten single deletions. The motivation is explained in the comment above class WBWIMemTable definition. The rest of the changes in WBWIMemTable are moving the iterator definition around.

Some intended follow ups:
1. support for merge operations
2. stats/logging around this option
3. tests improvement, including stress test support for the more comprehensive no_batched_op_stress.

Pull Request resolved: https://github.com/facebook/rocksdb/pull/13144

Test Plan:
* added new unit tests
* enabled in multi_ops_txns_stress test
* Benchmark: applying the change in 8222c0cafc4c6eb3a0d05807f7014b44998acb7a, I tested txn size of 10k and check perf context for write_memtable_time, write_wal_time and key_lock_wait_time(repurposed for transaction unlock time). Though the benchmark result number can be flaky, this shows memtable write time improved a lot (more than 100 times). The benchmark also shows that the remaining commit latency is from transaction unlock.
```
./db_bench --benchmarks=fillrandom --seed=1727376962 --threads=1 --disable_auto_compactions=1 --max_write_buffer_number=100 --min_write_buffer_number_to_merge=100 --writes=100000 --batch_size=10000 --transaction_db=1 --perf_level=4 --enable_pipelined_write=false --commit_bypass_memtable=1

commit_bypass_memtable = false
fillrandom   :       3.982 micros/op 251119 ops/sec 0.398 seconds 100000 operations;   27.8 MB/s PERF_CONTEXT:
write_memtable_time = 116950422
write_wal_time =      8535565
txn unlock time =     32979883

commit_bypass_memtable = true
fillrandom   :       2.627 micros/op 380559 ops/sec 0.263 seconds 100000 operations;   42.1 MB/s PERF_CONTEXT:
write_memtable_time = 740784
write_wal_time =      11993119
txn unlock time =     21735685
```

Reviewed By: jowlyzhang

Differential Revision: D66307632

Pulled By: cbi42

fbshipit-source-id: 6619af58c4c537aed1f76c4a7e869fb3f5098999
2024-12-05 15:00:17 -08:00
..
backup_engine_options.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
backupenginejni.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
cache.cc Put Cache and CacheWrapper in new public header (#11192) 2023-02-09 12:12:02 -08:00
cassandra_compactionfilterjni.cc Fix pointer to jlong conversion in 32 bit OS (#9396) 2022-03-01 09:02:15 -08:00
cassandra_value_operator.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
checkpoint.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
clock_cache.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
columnfamilyhandle.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
compact_range_options.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
compaction_filter.cc Run format check for *.h and *.cc files under java/ (#10851) 2022-10-25 09:26:51 -07:00
compaction_filter_factory.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
compaction_filter_factory_jnicallback.cc Run format check for *.h and *.cc files under java/ (#10851) 2022-10-25 09:26:51 -07:00
compaction_filter_factory_jnicallback.h Run format check for *.h and *.cc files under java/ (#10851) 2022-10-25 09:26:51 -07:00
compaction_job_info.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
compaction_job_stats.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
compaction_options.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
compaction_options_fifo.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
compaction_options_universal.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
comparator.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
comparatorjnicallback.cc Fix Java API ComparatorOptions use after delete error (#11176) 2023-02-17 13:03:41 -08:00
comparatorjnicallback.h Fix Java API ComparatorOptions use after delete error (#11176) 2023-02-17 13:03:41 -08:00
compression_options.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
concurrent_task_limiter.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
config_options.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
cplusplus_to_java_convert.h Fix pointer to jlong conversion in 32 bit OS (#9396) 2022-03-01 09:02:15 -08:00
env.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
env_options.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
event_listener.cc Fix pointer to jlong conversion in 32 bit OS (#9396) 2022-03-01 09:02:15 -08:00
event_listener_jnicallback.cc Add (& fix) some simple source code checks (#8821) 2021-09-07 21:19:27 -07:00
event_listener_jnicallback.h Add (& fix) some simple source code checks (#8821) 2021-09-07 21:19:27 -07:00
export_import_files_metadatajni.cc Add jni Support for API CreateColumnFamilyWithImport (#11646) 2023-11-06 07:38:42 -08:00
filter.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
hyper_clock_cache.cc Add HyperClockCache Java API. (#12065) 2023-11-16 15:46:31 -08:00
import_column_family_options.cc Add jni Support for API CreateColumnFamilyWithImport (#11646) 2023-11-06 07:38:42 -08:00
ingest_external_file_options.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
iterator.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
jni_multiget_helpers.cc JNI get_helper code sharing / multiGet() use efficient batch C++ support (#12344) 2024-03-12 12:42:08 -07:00
jni_multiget_helpers.h JNI get_helper code sharing / multiGet() use efficient batch C++ support (#12344) 2024-03-12 12:42:08 -07:00
jni_perf_context.cc Implement PerfContex#toString for the Java API (#12473) 2024-04-03 14:33:31 -07:00
jnicallback.cc Run format check for *.h and *.cc files under java/ (#10851) 2022-10-25 09:26:51 -07:00
jnicallback.h Run format check for *.h and *.cc files under java/ (#10851) 2022-10-25 09:26:51 -07:00
kv_helper.h Fix deprecated dynamic exception in internal_repo_rocksdb/repo/java/rocksjni/kv_helper.h +1 2024-06-13 12:41:13 -07:00
loggerjnicallback.cc Add native logger support to RocksJava (#12213) 2024-01-17 17:51:36 -08:00
loggerjnicallback.h Run format check for *.h and *.cc files under java/ (#10851) 2022-10-25 09:26:51 -07:00
lru_cache.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
memory_util.cc Run format check for *.h and *.cc files under java/ (#10851) 2022-10-25 09:26:51 -07:00
memtablejni.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
merge_operator.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
native_comparator_wrapper_test.cc Fix pointer to jlong conversion in 32 bit OS (#9396) 2022-03-01 09:02:15 -08:00
optimistic_transaction_db.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
optimistic_transaction_options.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
options.cc use nullptr instead of NULL / 0 in rocksdbjni (#12575) 2024-05-21 12:56:07 -07:00
options_util.cc java API - load block based table config (#10826) 2023-10-12 09:39:01 -07:00
persistent_cache.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
portal.h Add ticker stats for read corruption retries (#12923) 2024-08-12 15:32:07 -07:00
ratelimiterjni.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
remove_emptyvalue_compactionfilterjni.cc Fix pointer to jlong conversion in 32 bit OS (#9396) 2022-03-01 09:02:15 -08:00
restorejni.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
rocks_callback_object.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
rocksdb_exception_test.cc Run format check for *.h and *.cc files under java/ (#10851) 2022-10-25 09:26:51 -07:00
rocksjni.cc JNI get_helper code sharing / multiGet() use efficient batch C++ support (#12344) 2024-03-12 12:42:08 -07:00
slice.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
snapshot.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
sst_file_manager.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
sst_file_reader_iterator.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
sst_file_readerjni.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
sst_file_writerjni.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
sst_partitioner.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
statistics.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
statisticsjni.cc Fix tabs and lint-ignores (#6734) 2020-04-20 11:39:31 -07:00
statisticsjni.h Remove 'virtual' when implied by 'override' (#12319) 2024-01-31 13:14:42 -08:00
stderr_logger.cc Add native logger support to RocksJava (#12213) 2024-01-17 17:51:36 -08:00
table.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
table_filter.cc Fix pointer to jlong conversion in 32 bit OS (#9396) 2022-03-01 09:02:15 -08:00
table_filter_jnicallback.cc Run format check for *.h and *.cc files under java/ (#10851) 2022-10-25 09:26:51 -07:00
table_filter_jnicallback.h Run format check for *.h and *.cc files under java/ (#10851) 2022-10-25 09:26:51 -07:00
table_properties_collector_factory.cc Fix header files to meet Open source requirements (#12164) 2023-12-19 13:43:17 -08:00
table_properties_collector_factory.h Fix header files to meet Open source requirements (#12164) 2023-12-19 13:43:17 -08:00
testable_event_listener.cc Fix build on OpenBSD i386 (#12142) 2024-01-29 16:19:59 -08:00
thread_status.cc Run format check for *.h and *.cc files under java/ (#10851) 2022-10-25 09:26:51 -07:00
trace_writer.cc Run format check for *.h and *.cc files under java/ (#10851) 2022-10-25 09:26:51 -07:00
trace_writer_jnicallback.cc Run format check for *.h and *.cc files under java/ (#10851) 2022-10-25 09:26:51 -07:00
trace_writer_jnicallback.h Run format check for *.h and *.cc files under java/ (#10851) 2022-10-25 09:26:51 -07:00
transaction.cc JNI get_helper code sharing / multiGet() use efficient batch C++ support (#12344) 2024-03-12 12:42:08 -07:00
transaction_db.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
transaction_db_options.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
transaction_log.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
transaction_notifier.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
transaction_notifier_jnicallback.cc Run format check for *.h and *.cc files under java/ (#10851) 2022-10-25 09:26:51 -07:00
transaction_notifier_jnicallback.h Run format check for *.h and *.cc files under java/ (#10851) 2022-10-25 09:26:51 -07:00
transaction_options.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
ttl.cc use nullptr instead of NULL / 0 in rocksdbjni (#12575) 2024-05-21 12:56:07 -07:00
wal_filter.cc Run format check for *.h and *.cc files under java/ (#10851) 2022-10-25 09:26:51 -07:00
wal_filter_jnicallback.cc Run format check for *.h and *.cc files under java/ (#10851) 2022-10-25 09:26:51 -07:00
wal_filter_jnicallback.h Run format check for *.h and *.cc files under java/ (#10851) 2022-10-25 09:26:51 -07:00
write_batch.cc Replace ScopedArenaIterator with ScopedArenaPtr<InternalIterator> (#12470) 2024-03-22 13:40:42 -07:00
write_batch_test.cc Introduce a transaction option to skip memtable write during commit (#13144) 2024-12-05 15:00:17 -08:00
write_batch_with_index.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
write_buffer_manager.cc Change Java native methods to static (#11882) 2024-01-25 12:36:30 -08:00
writebatchhandlerjnicallback.cc Run format check for *.h and *.cc files under java/ (#10851) 2022-10-25 09:26:51 -07:00
writebatchhandlerjnicallback.h Run format check for *.h and *.cc files under java/ (#10851) 2022-10-25 09:26:51 -07:00