VictoriaMetrics/lib/storage
Zakhar Bessarab 8198e7241d
lib/mergeset: add sparse indexdb cache (#7269)
Related issue:
https://github.com/VictoriaMetrics/VictoriaMetrics/issues/7182

- add a separate index cache for searches which might read through large
amounts of random entries. Primary use-case for this is retention and
downsampling filters, when applying filters background merge needs to
fetch large amount of random entries which pollutes an index cache.
Using different caches allows to reduce effect on memory usage and cache
efficiency of the main cache while still having high cache hit rate. A
separate cache size is 5% of allowed memory.

- reduce size of indexdb/dataBlocks cache in order to free memory for
new sparse cache. Reduced size by 5% and moved this to a separate cache.

- add a separate metricName search which does not cache metric names -
this is needed in order to allow disabling metric name caching when
applying downsampling/retention filters. Applying filters during
background merge accesses random entries, this fills up cache and does
not provide an actual improvement due to random access nature.

Merge performance and memory usage stats before and after the change:

- before

![image](https://github.com/user-attachments/assets/485fffbb-c225-47ae-b5c5-bc8a7c57b36e)

- after

![image](https://github.com/user-attachments/assets/f4ba3440-7c1c-4ec1-bc54-4d2ab431eef5)

---------

Signed-off-by: Zakhar Bessarab <z.bessarab@victoriametrics.com>
(cherry picked from commit 837d0d136d)
2024-10-24 12:43:06 -03:00
..
block.go lib/encoding: optimize UnmarshalVarUint64, UnmarshalVarInt64 and UnmarshalBytes a bit 2024-05-14 01:30:25 +02:00
block_header.go lib/encoding: optimize UnmarshalVarUint64, UnmarshalVarInt64 and UnmarshalBytes a bit 2024-05-14 01:30:25 +02:00
block_header_test.go lib/storage: typo fix: umarshal -> unmarshal 2021-03-02 20:48:44 +02:00
block_stream_merger.go all: consistently use 'any' instead of 'interface{}' 2024-07-10 00:23:26 +02:00
block_stream_reader.go lib/storage: avoid superflouos copy of block header data 2024-02-23 01:11:31 +02:00
block_stream_reader_test.go Makefile: update golangci-lint from v1.51.2 to v1.54.2 2023-09-01 10:25:49 +02:00
block_stream_reader_timing_test.go lib/{storage,mergeset}: convert InitFromFilePart to MustInitFromFilePart 2023-04-14 15:47:20 -07:00
block_stream_writer.go lib/storage: consistently use atomic.* types instead of atomic.* function calls on ordinary types 2024-02-24 00:33:07 +02:00
block_stream_writer_timing_test.go lib/storage: consistently use atomic.* types instead of atomic.* function calls on ordinary types 2024-02-24 00:33:07 +02:00
block_test.go all: consistently use %w instead of %s in when error is passed to fmt.Errorf() 2023-10-26 09:44:40 +02:00
dedup.go docs: mention staleNaN handling during deduplication 2024-01-16 20:11:45 +02:00
dedup_test.go lib/storage: follow-up after 7c0ae3a86a 2022-12-08 18:18:49 -08:00
dedup_timing_test.go lib/storage: follow-up after 7c0ae3a86a 2022-12-08 18:18:49 -08:00
filenames.go lib/storage: consistently use OS-independent separator in file paths 2023-03-25 14:34:36 -07:00
index_db.go lib/mergeset: add sparse indexdb cache (#7269) 2024-10-24 12:43:06 -03:00
index_db_test.go lib/storage: restore ability to put empty metric ID list into tagFiltersToMetricIDsCache (#7064) 2024-09-20 17:38:56 +02:00
index_db_timing_test.go lib/storage: restore ability to put empty metric ID list into tagFiltersToMetricIDsCache (#7064) 2024-09-20 17:38:56 +02:00
inmemory_part.go lib/{storage,mergeset}: convert InitFromFilePart to MustInitFromFilePart 2023-04-14 15:47:20 -07:00
inmemory_part_test.go lib/{storage,mergeset}: convert InitFromFilePart to MustInitFromFilePart 2023-04-14 15:47:20 -07:00
inmemory_part_timing_test.go lib/storage: use deterministic random generator in tests 2023-01-23 20:12:32 -08:00
merge.go all: consistently use 'any' instead of 'interface{}' 2024-07-10 00:23:26 +02:00
merge_test.go lib/storage: consistently use atomic.* types instead of atomic.* function calls on ordinary types 2024-02-24 00:33:07 +02:00
merge_timing_test.go lib/storage: consistently use atomic.* types instead of atomic.* function calls on ordinary types 2024-02-24 00:33:07 +02:00
metaindex_row.go all: subsitute ioutil.ReadAll with io.ReadAll 2022-08-22 00:16:04 +03:00
metaindex_row_test.go lib/storage: correctly use maxBlockSize in various checks 2020-09-24 18:13:15 +03:00
metric_name.go lib/storage: change default value for maxLabelValueLen to 1024 (#6313) 2024-05-22 21:55:21 +02:00
metric_name_test.go app/vminsert: add support for data ingestion via other vminsert nodes 2021-05-08 19:53:45 +03:00
part.go lib/{storage,mergeset}: convert InitFromFilePart to MustInitFromFilePart 2023-04-14 15:47:20 -07:00
part_header.go lib/storage: add ability to use downsampling for the given series filter (#733) 2024-04-03 02:38:37 +03:00
part_search.go lib/storage: optimize partSearch.searchBHS() for common case when the TSID for the current block header is bigger or equal to the current tsid 2022-12-19 10:31:39 -08:00
part_search_test.go lib/{storage,mergeset}: convert InitFromFilePart to MustInitFromFilePart 2023-04-14 15:47:20 -07:00
part_search_timing_test.go all: consistently use %w instead of %s in when error is passed to fmt.Errorf() 2023-10-26 09:44:40 +02:00
partition.go Revert "refactor(vmstorage): Refactor the code to reduce the time complexity of MustAddRows and improve readability (#6629)" 2024-07-25 14:43:00 +02:00
partition_search.go all: consistently use 'any' instead of 'interface{}' 2024-07-10 00:23:26 +02:00
partition_search_test.go lib/storage: do not keep rawRows buffer across flush() calls 2024-02-23 01:06:09 +02:00
partition_test.go lib/storage: stop exposing vm_merge_need_free_disk_space metric 2023-09-25 17:00:14 +02:00
raw_block.go all: open-sourcing single-node version 2019-05-23 00:18:06 +03:00
raw_row.go lib/storage: consistently use atomic.* types instead of atomic.* function calls on ordinary types 2024-02-24 00:33:07 +02:00
search.go lib/storage: properly unmarshal SearchQuery (#7277) 2024-10-17 11:45:27 +02:00
search_test.go vmselect: add support of multi-tenant queries (#6346) 2024-10-01 16:37:18 +02:00
storage.go (app|lib)/vmstorage: do not increment vm_rows_ignored_total on NaNs (#7166) 2024-10-02 12:43:13 +02:00
storage_test.go lib/storage: Fix flaky test: TestStorageRotateIndexDB (#7267) 2024-10-24 15:45:54 +02:00
storage_timing_test.go Fix inconsistent error handling in Storage.AddRows() (#6583) 2024-07-17 12:55:07 +02:00
table.go Revert "refactor(vmstorage): Refactor the code to reduce the time complexity of MustAddRows and improve readability (#6629)" 2024-07-25 14:43:00 +02:00
table_search.go all: consistently use 'any' instead of 'interface{}' 2024-07-10 00:23:26 +02:00
table_search_test.go lib/storage: properly free up resources from newTestStorage() by calling stopTestStorage() 2023-07-13 17:13:34 -07:00
table_search_timing_test.go lib/storage: replace the remaining atomic.* functions with atomic.* types for the sake of consistency 2024-02-24 00:51:03 +02:00
table_test.go flagutil: Make .Msecs private (#4906) 2023-09-03 10:37:57 +02:00
table_timing_test.go lib/storage: properly free up resources from newTestStorage() by calling stopTestStorage() 2023-07-13 17:13:34 -07:00
tag_filters.go lib/logstorage: work-in-progress 2024-05-24 03:07:07 +02:00
tag_filters_test.go lib/logstorage: work-in-progress 2024-05-24 03:07:07 +02:00
tag_filters_timing_test.go all: fix golangci-lint(revive) warnings after 0c0ed61ce7 2024-04-03 03:00:45 +03:00
time.go app,lib: fix typos in comments (#3804) 2023-02-13 09:32:35 -08:00
time_test.go all: open-sourcing single-node version 2019-05-23 00:18:06 +03:00
tsid.go lib/storage: optimize TSID comparison 2019-09-26 14:20:02 +03:00
tsid_test.go all: open-sourcing cluster version 2019-05-23 00:25:38 +03:00