[ https://issues.apache.org/jira/browse/BEAM-5987?focusedWorklogId=168327&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-168327 ]
ASF GitHub Bot logged work on BEAM-5987: ---------------------------------------- Author: ASF GitHub Bot Created on: 21/Nov/18 15:58 Start Date: 21/Nov/18 15:58 Worklog Time Spent: 10m Work Description: dmvk commented on a change in pull request #7091: [BEAM-5987] Spark: Share cached side inputs between tasks. URL: https://github.com/apache/beam/pull/7091#discussion_r235445177 ########## File path: runners/spark/src/main/java/org/apache/beam/runners/spark/translation/MultiDoFnFunction.java ########## @@ -61,6 +66,17 @@ public class MultiDoFnFunction<InputT, OutputT> implements PairFlatMapFunction<Iterator<WindowedValue<InputT>>, TupleTag<?>, WindowedValue<?>> { + private static final Logger LOG = LoggerFactory.getLogger(MultiDoFnFunction.class); + + /** JVM wide side input cache. */ + private static final Map<String, CachedSideInputReader> sideInputReaders = + Collections.synchronizedMap(new WeakHashMap<>()); + + /** + * Id that is consistent among executors. We can not use stepName because of possible collisions. Review comment: After deserialization on the executor side ---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org Issue Time Tracking ------------------- Worklog Id: (was: 168327) Time Spent: 1h 40m (was: 1.5h) > Spark SideInputReader performance > --------------------------------- > > Key: BEAM-5987 > URL: https://issues.apache.org/jira/browse/BEAM-5987 > Project: Beam > Issue Type: Bug > Components: runner-spark > Affects Versions: 2.8.0 > Reporter: David Moravek > Assignee: David Moravek > Priority: Major > Fix For: 2.9.0 > > Attachments: Screen Shot 2018-11-06 at 13.05.36.png > > Time Spent: 1h 40m > Remaining Estimate: 0h > > We did some profiling of a spark job and 90% of the application time was > spent on side input deserialization. > For spark, an easy fix is to cache materialized side inputs per bundle. This > improved running time of the profiled job from 3 hours to 30 minutes. -- This message was sent by Atlassian JIRA (v7.6.3#76005)