[
https://issues.apache.org/jira/browse/HBASE-17738?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16055602#comment-16055602
]
Anoop Sam John commented on HBASE-17738:
----------------------------------------
Some unwanted code is there in the latest patch which checks the Max direct
buffer size etc. Pls clean and avoid unused imports in the patch. Ya I agree
to this approach alone (After said the issues with fully Unsafe based creation
of DBBs)
We should create the new Callable this way that it will create N number of BBs
are return us? No need to submit task for each of the BB create. Say we have
99 BBs to be created and we have 10 threads, make 9 threads to create 10 BBs
each and the last one to create 9? This might be more efficient than giving
99 Callables and then each been taken up by random threads?
> BucketCache startup is slow
> ---------------------------
>
> Key: HBASE-17738
> URL: https://issues.apache.org/jira/browse/HBASE-17738
> Project: HBase
> Issue Type: Sub-task
> Components: BucketCache
> Affects Versions: 2.0.0
> Reporter: stack
> Assignee: ramkrishna.s.vasudevan
> Fix For: 2.0.0
>
> Attachments: HBASE-17738_2.patch, HBASE-17738_2.patch,
> HBASE-17738_3.patch, HBASE-17738_4.patch, HBASE-17738_5_withoutUnsafe.patch,
> HBASE-17738.patch
>
>
> If you set bucketcache size at 64G say and then start hbase, it takes a long
> time. Can we do the allocations in parallel and not inline with the server
> startup?
> Related, prefetching on a bucketcache is slow. Speed it up.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
