Commit Graph

130 Commits

Author SHA1 Message Date
Aliaksandr Valialkin
6f61fd367a lib/storage: add BenchmarkHeadPostingForMatchers similar to the benchmark from Prometheus
See the corresponding benchmark in Prometheus - 23c0299d85/tsdb/head_bench_test.go (L52)

The benchmark allows performing apples-to-apples comparison of time series search
in Prometheus and VictoriaMetrics. The following article - https://www.robustperception.io/evaluating-performance-and-correctness -
contains incorrect numbers for VictoriaMetrics, since there wasn't this benchmark yet. Fix it.

Benchmarks can be repeated with the following commands from Prometheus and VictoriaMetrics source code roots:

- Prometheus: GOMAXPROCS=1 go test ./tsdb/ -run=111 -bench=BenchmarkHeadPostingForMatchers
- VictoriaMetrics: GOMAXPROCS=1 go test ./lib/storage/ -run=111 -bench=BenchmarkHeadPostingForMatchers

Benchmark results:
benchmark                                                          old ns/op      new ns/op     delta
BenchmarkHeadPostingForMatchers/n="1"                              272756688      364977        -99.87%
BenchmarkHeadPostingForMatchers/n="1",j="foo"                      138132923      1181636       -99.14%
BenchmarkHeadPostingForMatchers/j="foo",n="1"                      134723762      1141578       -99.15%
BenchmarkHeadPostingForMatchers/n="1",j!="foo"                     195823953      1148056       -99.41%
BenchmarkHeadPostingForMatchers/i=~".*"                            7962582919     8716755       -99.89%
BenchmarkHeadPostingForMatchers/i=~".+"                            7589543864     12096587      -99.84%
BenchmarkHeadPostingForMatchers/i=~""                              1142371741     16164560      -98.59%
BenchmarkHeadPostingForMatchers/i!=""                              9964150263     12230021      -99.88%
BenchmarkHeadPostingForMatchers/n="1",i=~".*",j="foo"              216995884      1173476       -99.46%
BenchmarkHeadPostingForMatchers/n="1",i=~".*",i!="2",j="foo"       202541348      1299743       -99.36%
BenchmarkHeadPostingForMatchers/n="1",i!=""                        486285711      11555193      -97.62%
BenchmarkHeadPostingForMatchers/n="1",i!="",j="foo"                350776931      5607506       -98.40%
BenchmarkHeadPostingForMatchers/n="1",i=~".+",j="foo"              380888565      6380335       -98.32%
BenchmarkHeadPostingForMatchers/n="1",i=~"1.+",j="foo"             89500296       2078970       -97.68%
BenchmarkHeadPostingForMatchers/n="1",i=~".+",i!="2",j="foo"       379529654      6561368       -98.27%
BenchmarkHeadPostingForMatchers/n="1",i=~".+",i!~"2.*",j="foo"     424563825      6757132       -98.41%

The first column (old) is for Prometheus, the second column (new) is for VictoriaMetrics.

Prometheus was using 3.5GB of RAM during the benchmark, while VictoriaMetrics was using 400MB of RAM.
2019-11-18 18:47:02 +02:00
Aliaksandr Valialkin
d297b65089 lib/storage: add vm_cache_size_bytes{type="storage/hour_metric_ids"} metric 2019-11-13 20:26:05 +02:00
Aliaksandr Valialkin
494ad0fdb3 lib/storage: remove inmemory index for recent hour, since it uses too much memory
Production workload shows that the index requires ~4Kb of RAM per active time series.
This is too much for high number of active time series, so let's delete this index.

Now the queries should fall back to the index for the current day instead of the index
for the recent hour. The query performance for the current day index should be good enough
given the 100M rows/sec scan speed per CPU core.
2019-11-13 18:08:58 +02:00
Aliaksandr Valialkin
633dd81bb5 lib/storage: add -disableRecentHourIndex flag for disabling inmemory index for recent hour
This may be useful for saving RAM on high number of time series aka high cardinality
2019-11-13 15:10:12 +02:00
Aliaksandr Valialkin
f1620ba7c0 lib/storage: fix inmemory inverted index issues found in v1.29
Issues fixed:
- Slow startup times. Now the index is loaded from cache during start.
- High memory usage related to superflouos index copies every 10 seconds.
2019-11-13 13:35:38 +02:00
Mike Poindexter
955a592106 Add test for invalid caching of tsids (#232)
* Add test for invalid caching of tsids

* Clean up error handling
2019-11-12 15:52:46 +02:00
Oleg Kovalov
74ba42d111 fix misspelled words (#229) 2019-11-12 00:18:24 +02:00
Aliaksandr Valialkin
c48e39eea9 lib/storage: add tests for dateMetricIDCache 2019-11-11 13:21:05 +02:00
Aliaksandr Valialkin
6bdde0d6d4 lib/storage: eliminate data race when updating lastSyncTime in dateMetricIDCache.Has 2019-11-10 22:04:23 +02:00
Aliaksandr Valialkin
9ea2bd822e lib/storage: implement per-day inverted index 2019-11-10 00:20:32 +02:00
Aliaksandr Valialkin
dea2f3efed lib/storage: use specialized cache for (date, metricID) entries
This improves ingestion performance.
2019-11-09 23:09:18 +02:00
Aliaksandr Valialkin
9a43902bd8 lib/storage: remove unused code from getMetricIDsForTimeRange: it is expected that time range is always non-zero 2019-11-09 19:03:51 +02:00
Aliaksandr Valialkin
c16e17dede lib/storage: properly set time range when deleting time series 2019-11-09 18:50:02 +02:00
Aliaksandr Valialkin
8126007c15 lib/storage: obtain all the time series ids from (tag->metricIDs) rows instead of (metricID->TSID) rows, since this much faster 2019-11-09 18:04:26 +02:00
Aliaksandr Valialkin
50773348d3 lib/storage: small code prettifying 2019-11-09 14:01:24 +02:00
Aliaksandr Valialkin
0bc54c23ce lib/storage: inmemoryInvertedIndex prettifying 2019-11-09 14:01:24 +02:00
Aliaksandr Valialkin
46e67bb78c lib/storage: export vm_new_timeseries_created_total metric for determining time series churn rate 2019-11-08 19:58:21 +02:00
Aliaksandr Valialkin
0063c857f5 lib/storage: add inmemory inverted index for the last hour
It should improve performance for `last N hours` dashboards with update intervals smaller than 1 hour.
2019-11-08 19:37:46 +02:00
Aliaksandr Valialkin
89c03a5464 lib/storage: populate partition names from both small and big directories
Certain partition directories may be missing after restoring from backups
if they had no data. Re-create such directories on start.
2019-11-06 19:50:21 +02:00
Aliaksandr Valialkin
1c777e0245 lib/storage: substitute error message about unsorted items in the index block after metricIDs merge with counter
The origin of the error has been detected and documented in the code,
so it is enough to export a counter for such errors at `vm_index_blocks_with_metric_ids_incorrect_order_total`,
so it could be monitored and alerted on high error rates.

Export also the counter for processed index blocks with metricIDs - `vm_index_blocks_with_metric_ids_processed_total`,
so its' rate could be compared to `rate(vm_index_blocks_with_metric_ids_incorrect_order_total)`.
2019-11-06 14:32:41 +02:00
Aliaksandr Valialkin
c567a4353a lib/storage: take into account the requested time range when caching TSIDs for the given tag filters 2019-11-06 14:32:41 +02:00
Aliaksandr Valialkin
c6564c5d26 lib/storage: dump incorrectly sorted items on a single line; this should simplify error reporting 2019-11-05 18:41:50 +02:00
Aliaksandr Valialkin
a10c4cad85 lib/storage: return back finalPartsToMerge from 2 to 3 in order to prevent from excessive merges in old partitions 2019-11-05 17:28:57 +02:00
Aliaksandr Valialkin
e5b1fa0c38 lib/storage: separate the max inverted index scan loops per metric into fast and slow loops
Slow loops could require seeks and expensive regexp matching, while fast loops just scans
all the metricIDs for the given `tag=value` prefix. So these operations must have separate
max loops multiplier.
2019-11-05 17:28:57 +02:00
Aliaksandr Valialkin
f93c4f2493 lib/storage: skip repeated useless work when intersection of metricIDs with the given filter is too expensive
This should improve performance for query filters over big number of time series.
2019-11-05 14:35:55 +02:00
Aliaksandr Valialkin
f48e97263c lib/storage: reduce the maximum inverted index scans before giving up to label filters matching by metric name
The new value reduces the amount of wasted work during index scans over big number of time series.
2019-11-05 14:35:53 +02:00
Aliaksandr Valialkin
d2f688c550 lib/storage: try potentially faster tag filters at first, then apply slower tag filters
The fastest tag filters are non-negative non-regexp, since they are the most specific.
The slowest tag filters are negative regexp, since they require scanning
all the entries for the given label.
2019-11-05 14:35:48 +02:00
Aliaksandr Valialkin
2a38d30f93 lib/storage: pass pointer to MetricName in Fatalf, so it is properly detected as an interface with String() method
This fixes lint errors
2019-11-04 01:06:45 +02:00
Artem Navoiev
e05500cbd4 add unittests for bytesutil and storage (#221) 2019-11-04 00:57:24 +02:00
Aliaksandr Valialkin
f5fbc3ffd7 lib/{storage,uint64set}: add Set.Union() function and use it 2019-11-04 00:48:32 +02:00
Aliaksandr Valialkin
23e078261e lib/storage: tune the returned value from adjustMaxMetricsAdaptive 2019-11-04 00:45:28 +02:00
Aliaksandr Valialkin
386c349c8c lib/storage: remove interface conversion in hot path during block merging
This should improve merge speed a bit for parts with big number of small blocks.
2019-11-03 12:33:48 +02:00
Aliaksandr Valialkin
26ffc77622 lib/{storage,mergeset}: create missing partition directories after restoring from backups
Backup tools could skip empty directories. So re-create such directories on the first run.
2019-11-02 02:27:19 +02:00
Aliaksandr Valialkin
6ab9c98a1e app/vmstorage: add -bigMergeConcurrency and -smallMergeConcurrency flags for tuning the maximum number of CPU cores used during merges 2019-10-31 16:17:29 +02:00
Aliaksandr Valialkin
6a22727676 lib/storage: optimize getMetricIDsForRecentHours for per-tenant lookups 2019-10-31 15:51:09 +02:00
Aliaksandr Valialkin
ca480915ca lib/storage: small cleanup in Storage.add 2019-10-31 14:30:22 +02:00
Aliaksandr Valialkin
a27e034a40 lib/storage: get parts to merge after applying the limit on the number of concurrent merges
This should reduce write amplification under high ingestion rate.
2019-10-30 00:09:44 +02:00
Aliaksandr Valialkin
5d2276dbf7 lib/{mergeset,storage}: limit the maximum number of concurrent merges; leave smaller number of parts during final merge 2019-10-29 12:45:37 +02:00
hanzai
52778da1f3 warns during rows addition (#214) 2019-10-20 23:38:51 +03:00
Aliaksandr Valialkin
78fc35c9b1 all: make fmt 2019-10-17 20:05:12 +03:00
Aliaksandr Valialkin
5b01b7fb01 all: add support for GOARCH=386 and fix all the issues related to 32-bit architectures such as GOARCH=arm
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/212
2019-10-17 18:27:49 +03:00
Aliaksandr Valialkin
661b8ede5b lib/storage: harden the check that the original items are sorted after mergeTagToMetricIDsRows fails to preserve sort order 2019-10-09 12:13:43 +03:00
Aliaksandr Valialkin
7f4a04ee6a lib/storage: typo fix in comment to maxRowsPerSmallPart. 2019-10-08 18:51:56 +03:00
Aliaksandr Valialkin
7e410e1412 lib/storage: add tests for mergeTagToMetricIDsRows and return the original items if the function breaks items` ordering.
This should save from data corruption issues revealed in the previous releases up to v1.28.0-beta5.
2019-10-08 16:35:39 +03:00
Aliaksandr Valialkin
d0b4590099 lib/storage: optimize TSID comparison 2019-09-26 14:20:02 +03:00
Aliaksandr Valialkin
95e3d648cb lib/storage: verify whether items are sorted in the end of call to mergeTagToMetricIDsRows
This should prevent from inverted index corruption if bug in mergeTagToMetricIDsRows is discovered.
2019-09-26 13:13:58 +03:00
Aliaksandr Valialkin
2b8358726f lib/storage: properly match labels against regexp with (?i) flag
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/161
2019-09-26 11:03:26 +03:00
Aliaksandr Valialkin
4e3871ac1e lib/storage: add missing break in removeDuplicateMetricIDs 2019-09-25 18:23:13 +03:00
Aliaksandr Valialkin
4468f9f966 lib/storage: remove duplicate MetricIDs in tag->metricIDs items before writing them into inverted index 2019-09-25 17:57:36 +03:00
Aliaksandr Valialkin
adc18c3ee6 lib/{mergeset,storage}: do not cache inverted index blocks containing tag->metricIDs items
This should reduce the amounts of used RAM during queries with filters over big number of time series.
2019-09-25 13:48:24 +03:00