You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@ignite.apache.org by "Ivan Pavlukhin (JIRA)" <ji...@apache.org> on 2019/01/14 12:32:00 UTC
[jira] [Comment Edited] (IGNITE-10754) Query history statistics API
[ https://issues.apache.org/jira/browse/IGNITE-10754?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16742005#comment-16742005 ]
Ivan Pavlukhin edited comment on IGNITE-10754 at 1/14/19 12:31 PM:
-------------------------------------------------------------------
[~jooger], see my comments at [GitHub|https://github.com/apache/ignite/pull/5805#pullrequestreview-192109814]. Most serious ones are:
1. Race condition in {{QueryHistoryTracker.touch}} method which might lead to history items which will stay in history forever (correct me if I am wrong).
2. CHM {{size}} method performance is not constant. It might lead to negative performance effects.
was (Author: pavlukhin):
[~jooger], see my comments at GitHub. Most serious ones are:
1. Race condition in {{QueryHistoryTracker.touch}} method which might lead to history items which will stay in history forever (correct me if I am wrong).
2. CHM {{size}} method performance is not constant. It might lead to negative performance effects.
> Query history statistics API
> ----------------------------
>
> Key: IGNITE-10754
> URL: https://issues.apache.org/jira/browse/IGNITE-10754
> Project: Ignite
> Issue Type: Task
> Components: sql
> Reporter: Yury Gerzhedovich
> Assignee: Yury Gerzhedovich
> Priority: Major
> Labels: iep-29, monitoring
> Time Spent: 40m
> Remaining Estimate: 0h
>
> As of now we have query statistics (*_org.apache.ignite.IgniteCache#queryMetrics_*) , but have few issues.
> 1) Duration execution it just time between start execution and return cursor to client and doesn't include all life time of query.
> 2) It doesn't know about multistatement queries. Such queries participate in statistics as single query without splitting.
> 3) API to access the statistics expose as depend on cache, however query don't have such dependency.
>
> Need to create parallel similar realization as we already have.
> Use new infrastructure of tracking running queries developed under IGNITE-10621 and update statistics on unregister phase.
> Expose API on upper level then it placed now. Right place will be written later.
>
>
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)