[
https://issues.apache.org/jira/browse/HBASE-17738?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
ramkrishna.s.vasudevan updated HBASE-17738:
-------------------------------------------
Attachment: HBASE-17738_2.patch
This patch uses the logic of creating bytebuffers as done by snazy/ohc.
It makes the RS start up even faster. The 150G BB itself takes 5 secs only but
the catch is that this does not honour the --XXDirectMemorySize param meter as
the allocation is just done assuming we have enough memory.
This is what snazy/ohc also makes use of.
So for 125G of offheap memory allocation
BEfore patch - 40 sec
With HBASE-17738.patch - 10 to 12 secs
With HBASE-17738_2.patch -5 to 6 secs.
> 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.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
(v6.3.15#6346)
