[ https://issues.apache.org/jira/browse/YARN-3942?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14906794#comment-14906794 ]
Li Lu commented on YARN-3942: ----------------------------- Thanks [~jlowe]! Unfortunately we haven't got a chance to test the latency for the new storage. As a reference, we can find something from leveldb's performance benchmark site (http://leveldb.googlecode.com/svn/trunk/doc/benchmark.html). For each of the random entity get/set we trigger two level db ops. For sequential reads (iterators) we generate at most one random read for the starting position, and then just sequential reads. I haven't tough caching policy in the first draft, but would definitely want to hear some feedbacks and suggestions from the community. > Timeline store to read events from HDFS > --------------------------------------- > > Key: YARN-3942 > URL: https://issues.apache.org/jira/browse/YARN-3942 > Project: Hadoop YARN > Issue Type: Improvement > Components: timelineserver > Reporter: Jason Lowe > Assignee: Jason Lowe > Attachments: YARN-3942-leveldb.001.patch, YARN-3942.001.patch > > > This adds a new timeline store plugin that is intended as a stop-gap measure > to mitigate some of the issues we've seen with ATS v1 while waiting for ATS > v2. The intent of this plugin is to provide a workable solution for running > the Tez UI against the timeline server on a large-scale clusters running many > thousands of jobs per day. -- This message was sent by Atlassian JIRA (v6.3.4#6332)