Jensen
Jensen
same problem, here is answer gen by copilot The error message indicates that the installation of the [`dtw`](command:_github.copilot.openSymbolFromReferences?%5B%22dtw%22%2C%5B%7B%22uri%22%3A%7B%22%24mid%22%3A1%2C%22fsPath%22%3A%22%2Fhome%2Fjensen%2FProjects%2Fquant%2Fevaluate-factor%2Fpkg%2Fframework%2Fregisters%2Fdtw.py%22%2C%22external%22%3A%22file%3A%2F%2F%2Fhome%2Fjensen%2FProjects%2Fquant%2Fevaluate-factor%2Fpkg%2Fframework%2Fregisters%2Fdtw.py%22%2C%22path%22%3A%22%2Fhome%2Fjensen%2FProjects%2Fquant%2Fevaluate-factor%2Fpkg%2Fframework%2Fregisters%2Fdtw.py%22%2C%22scheme%22%3A%22file%22%7D%2C%22pos%22%3A%7B%22line%22%3A5%2C%22character%22%3A5%7D%7D%5D%5D "Go to definition") package is failing because it cannot find the `imp`...
性能退化的job在跑`insert into table_with_com_pk_index_for_insert_1B`的时候, 从`2024 11:43:54`开始, 到`2024 14:12:41`结束 它的CPU截图如下所示 : *** 作为对比, 之前的jobs的CPU截图如下所示 : gc时间暴涨
```sh 9f1c2dfa9 fix delete page error and some typos (#17087)
> [0626] 1.2-dev: [2545ce4](https://github.com/matrixorigin/matrixone/commit/2545ce4c726d5c9e64f7a6b8307b5da0e6da950d) job url: https://github.com/matrixorigin/mo-nightly-regression/actions/runs/9660414495/job/26646571031 > > > profile: https://grafana.ci.matrixorigin.cn/explore?panes=%7B%22cg0%22:%7B%22datasource%22:%22pyroscope%22,%22queries%22:%5B%7B%22groupBy%22:%5B%5D,%22labelSelector%22:%22%7Bnamespace%3D%5C%22mo-big-data-20240625%5C%22%7D%22,%22queryType%22:%22both%22,%22refId%22:%22A%22,%22profileTypeId%22:%22memory:inuse_space:bytes:space:bytes%22,%22datasource%22:%7B%22type%22:%22grafana-pyroscope-datasource%22,%22uid%22:%22pyroscope%22%7D%7D%5D,%22range%22:%7B%22from%22:%221719342869000%22,%22to%22:%221719349741000%22%7D%7D%7D&schemaVersion=1&orgId=1 从pprof上看和main上性能退化是类似的
表因是最近改了`GOLIMIT`的参数, 根因还是在insert into select 的时候内存压力太大了, 导致疯狂gc - https://github.com/matrixorigin/mo-nightly-regression/commit/2726e7b3eab0ff8d4a5417fc063b3b5df42e5b78 继续做 - https://github.com/matrixorigin/matrixone/issues/16375
请魏璐帮忙跑了一下main上合并了fuzzy内存优化相关的代码, 可以看到gc的压力小了, 耗时回到了一小时以内. https://github.com/matrixorigin/mo-nightly-regression/actions/runs/9698664379/job/26775565464 profile link : https://grafana.ci.matrixorigin.cn/explore?panes=%7B%22cg0%22:%7B%22datasource%22:%22pyroscope%22,%22queries%22:%5B%7B%22groupBy%22:%5B%5D,%22labelSelector%22:%22%7Bnamespace%3D%5C%22mo-big-data-20240627%5C%22%7D%22,%22queryType%22:%22both%22,%22refId%22:%22A%22,%22profileTypeId%22:%22memory:inuse_space:bytes:space:bytes%22,%22datasource%22:%7B%22type%22:%22grafana-pyroscope-datasource%22,%22uid%22:%22pyroscope%22%7D%7D%5D,%22range%22:%7B%22from%22:%221719519314000%22,%22to%22:%221719522812000%22%7D%7D%7D&schemaVersion=1&orgId=1 cpu : inuse-memory :
但其实不能确定是不是由我的优化来fix的, 毕竟gc跟object数量有关系, 但fuzzy的内存消耗是很大头的, 这个跟gc的直接关系可能不大, 除了我的优化以外, 可能有助于修复这个issue的pr有 - 4f7b02b36 reduce memory allocation by refactoring process structure (#17183) - 8e7d832b8 malloc: optimizations (#17131) - 588707ecc fix memory leak in flushAObjsForSnapshot (#17124) 可能需要...
main上会继续跑big data query, 再观察一段时间