[
https://issues.apache.org/jira/browse/GEODE-6186?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16724414#comment-16724414
]
ASF subversion and git services commented on GEODE-6186:
--------------------------------------------------------
Commit ebde19f7e1a5839331161976b70c022a595be25c in geode's branch
refs/heads/develop from Barry Oglesby
[ https://gitbox.apache.org/repos/asf?p=geode.git;h=ebde19f ]
GEODE-6186: Reduced the number of EntryNotFoundExceptions thrown during wan
conflation
> Reduce the number of EntryNotFoundExceptions during AsyncEventQueue batch
> processing with conflation enabled
> ------------------------------------------------------------------------------------------------------------
>
> Key: GEODE-6186
> URL: https://issues.apache.org/jira/browse/GEODE-6186
> Project: Geode
> Issue Type: Bug
> Components: wan
> Reporter: Barry Oglesby
> Priority: Major
> Labels: pull-request-available
> Time Spent: 20m
> Remaining Estimate: 0h
>
> Reduce the number of EntryNotFoundExceptions during AsyncEventQueue batch
> processing with conflation enabled
> This test:
> 3000 iterations of putAlls with the same 1500 keys into a partitioned region
> attached to async-event-queue:
> <async-event-queue id="test" batch-size="100000" batch-time-interval="250"
> dispatcher-threads="1" parallel="true" enable-batch-conflation="true">
> Produces these numbers in the current code (4 different runs):
> {noformat}
> numBatches=645; numENFEs=8622196; totalPeekTime=178517; averagePeekTime=276;
> totalProcessBatchTime=38936; averageProcessBatchTime=60
> numBatches=660; numENFEs=8467986; totalPeekTime=182985; averagePeekTime=277;
> totalProcessBatchTime=34335; averageProcessBatchTime=52
> numBatches=646; numENFEs=8563364; totalPeekTime=179624; averagePeekTime=278;
> totalProcessBatchTime=37342; averageProcessBatchTime=57
> numBatches=632; numENFEs=8716942; totalPeekTime=175570; averagePeekTime=277;
> totalProcessBatchTime=39732; averageProcessBatchTime=62
> {noformat}
> After some changes mainly in BucketRegionQueue:
> {noformat}
> numBatches=782; numENFEs=3621039; totalPeekTime=195760; averagePeekTime=250;
> totalProcessBatchTime=18724; averageProcessBatchTime=23
> numBatches=791; numENFEs=3604933; totalPeekTime=197980; averagePeekTime=250;
> totalProcessBatchTime=18587; averageProcessBatchTime=23
> numBatches=790; numENFEs=3600038; totalPeekTime=197774; averagePeekTime=250;
> totalProcessBatchTime=18611; averageProcessBatchTime=23
> numBatches=795; numENFEs=3584490; totalPeekTime=199060; averagePeekTime=250;
> totalProcessBatchTime=18063; averageProcessBatchTime=22
> {noformat}
> numBatches is the number of batches peeked
> numENFEs is the number of EntryNotFoundExceptions thrown
> totalPeekTime is the total time to peek all batches
> averagePeekTime is the average time to peek a batch
> totalProcessBatchTime is the total time to process all batches
> averageProcessBatchTime is the average time to process a batch (includes
> listener callback and remove from queue)
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
