ClickHouse/docker/test/performance-comparison/config/config.d/zzz-perf-comparison-tweaks-config.xml
Azat Khuzhin 9e622b5add perf: switch *_log tables to Memory engine (attempt to reduce cache misses)
trace_log/query_log from performance tests shows (for
cases when prewarm query fails with timeout, 15sec) excessive
writeTraceInfo() in trace_log and QueryProfilerRuns in query_log, but
this is not the root cause of the timeout, but consequence.

Also query_log shows that on failures the following profile events has
significantly higher values:
- PerfLocalMemoryMisses (6.3x more)
- PerfLocalMemoryReferences (7x more)
- PerfDataTLBMisses (6.9x more)
- PerfInstructionTLBMisses (6.4x more)

During looking at performance tests logs I noticed that once the prewarm
query fails other server (left/right) was merging (MergeTree) something
in *_log tables.

But, using MergeTree for *_log in performance tests is useless, since
anyway environment for performance tests uses ramdrive.

And so MergeTree merges just increase overhead.

Eventually I expect that this should decrease extra memory referencing
and so this should decrease cache/TLB misses.

CI: https://clickhouse-test-reports.s3.yandex.net/30886/c504e0c08df7a926bb479a1d297f326f5c48a32f/performance_comparison/report.html#fail1

v2: <partition_by remove="remove"/>
2021-11-04 09:26:46 +03:00

53 lines
1.6 KiB
XML

<clickhouse>
<http_port remove="remove"/>
<mysql_port remove="remove"/>
<postgresql_port remove="remove"/>
<interserver_http_port remove="remove"/>
<tcp_with_proxy_port remove="remove"/>
<keeper_server remove="remove"/>
<zookeeper remove="remove"/>
<listen_host>::</listen_host>
<logger>
<console>true</console>
</logger>
<text_log remove="remove"/>
<crash_log remove="remove"/>
<query_views_log remove="remove"/>
<part_log remove="remove"/>
<opentelemetry_span_log remove="remove"/>
<session_log remove="remove"/>
<!-- performance tests does not uses real block devices,
instead they stores everything in memory.
And so, to avoid extra memory reference switch *_log to Memory engine. -->
<query_log>
<engine>ENGINE = Memory</engine>
<partition_by remove="remove"/>
</query_log>
<query_thread_log>
<engine>ENGINE = Memory</engine>
<partition_by remove="remove"/>
</query_thread_log>
<trace_log>
<engine>ENGINE = Memory</engine>
<partition_by remove="remove"/>
</trace_log>
<metric_log>
<engine>ENGINE = Memory</engine>
<partition_by remove="remove"/>
</metric_log>
<asynchronous_metric_log>
<engine>ENGINE = Memory</engine>
<partition_by remove="remove"/>
</asynchronous_metric_log>
<uncompressed_cache_size>1000000000</uncompressed_cache_size>
<asynchronous_metrics_update_period_s>10</asynchronous_metrics_update_period_s>
<remap_executable replace="replace">true</remap_executable>
</clickhouse>