[
https://issues.apache.org/jira/browse/HBASE-8420?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13656515#comment-13656515
]
Jimmy Xiang commented on HBASE-8420:
------------------------------------
bq. I think you have a leak if the client doesn't come back to get the
prefetched data, at least you should decrement globalPrefetchedResultSize in
RegionScannerHolder.closeScanner().
Good catch. I will fix this.
bq. 512MB seems big, I'd be more comfortable with 128MB.
How about 256 MB? 128MB is a little small.
bq. Is it possible to unit test this?
For the memory cap, it's probably hard to unit test. I am doing some manual
testing.
bq. Have you run a concurrent workload against capped pre-fetching?
Will set the cap to a small number and do some manual testing.
bq. Should we report on the state of globalPrefetchedResultSize? I'm
wondering what could be helpful to see when debugging scan slowness WRT this
rate limiting. Maybe a TRACE every time we skip a prefetch and output
totalPrefetchedResultSize?
Sure. Will do.
> Port HBASE-6874 Implement prefetching for scanners from 0.89-fb
> -----------------------------------------------------------------
>
> Key: HBASE-8420
> URL: https://issues.apache.org/jira/browse/HBASE-8420
> Project: HBase
> Issue Type: Improvement
> Reporter: Jimmy Xiang
> Assignee: Jimmy Xiang
> Attachments: 0.94-8420_v1.patch, 0.94-8420_v2.patch,
> trunk-8420_v1.patch, trunk-8420_v2.patch, trunk-8420_v3.patch
>
>
> This should help scanner performance. We should have it in trunk.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
