[
https://issues.apache.org/jira/browse/HBASE-17747?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15923524#comment-15923524
]
Yu Li commented on HBASE-17747:
-------------------------------
bq. Have you test it in the normal mode for HBase? Which GC algorithm do you
use?
Yes, please see [this
comment|https://issues.apache.org/jira/browse/HBASE-17747?focusedCommentId=15907134&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-15907134],
using CMS for GC.
bq. I'm a little worried about the memory pressure if you use soft reference
Regarding memory pressure, actually in embedded mode the pressure is much
higher than the normal(distributed) mode, and it's weak reference which causing
frequent full GC and soft reference fixed it.
bq. I used to use soft reference somewhere and got OOME...
This phenomenon seems to be against the design of soft reference, may I doubt
it was caused by some JVM implementation bug? (smile)
> Support both weak and soft object pool
> --------------------------------------
>
> Key: HBASE-17747
> URL: https://issues.apache.org/jira/browse/HBASE-17747
> Project: HBase
> Issue Type: Improvement
> Affects Versions: 2.0
> Reporter: Yu Li
> Assignee: Yu Li
> Fix For: 2.0
>
> Attachments: HBASE-17747.patch, HBASE-17747.v2.patch,
> HBASE-17747.v3.patch
>
>
> During YCSB testing on embedded mode after HBASE-17744, we found that under
> high read load GC is quite severe even with offheap L2 cache. After some
> investigation, we found it's caused by using weak reference in
> {{IdReadWriteLock}}. In embedded mode the read is so quick that the lock
> might already get promoted to the old generation when the weak reference is
> cleared, which causes dirty card table (old reference get removed and new
> lock object set into {{referenceCache}}, see {{WeakObjectPool#get}}) thus
> slowing YGC. In distributed mode there'll also be more lock object created
> with weak reference than soft reference that slowing down the processing.
> So we proposed to use soft reference for this {{IdReadWriteLock}} used in
> cache, which won't get cleared until JVM memory is not enough, and could
> resolve the issue mentioned above. What's more, we propose to extend the
> {{WeakObjectPool}} to be more generate to support both weak and soft
> reference.
> Note that the GC issue only emerges under embedded mode with DirectOperator,
> in which case all costs on the wire is removed thus produces extremely high
> concurrency.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
