You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@hbase.apache.org by "ramkrishna.s.vasudevan (JIRA)" <ji...@apache.org> on 2014/03/26 11:50:16 UTC

[jira] [Commented] (HBASE-10772) Use ByteRanges instead of ByteBuffers in BlockCache

    [ https://issues.apache.org/jira/browse/HBASE-10772?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13947793#comment-13947793 ] 

ramkrishna.s.vasudevan commented on HBASE-10772:
------------------------------------------------

Would like bring up a discussion here, can open a relevant JIRA for that.
HBASE-10771 adds some geXXX and putXXX type of APIs, put() will add a PBR itself. 
While working on BR inside BlockCache, the current way of BlockCache that use offheap - tries to slice the given DBB and uses a DBB buffer pool.  So if we need such use cases with BRs instead of DBB and those BRs should be offheap, how do we do that.  So should we add some APIs like slice, duplicate() in DBB to BRs and try reusing the allocated BRs? Currently we don't have any offheap based BR.  So if we try to create a BR that is backed by DBB and to ensure the current behaviour works of BR pools etc we may need such things in BR also. What do you think?

> Use ByteRanges instead of ByteBuffers in BlockCache
> ---------------------------------------------------
>
>                 Key: HBASE-10772
>                 URL: https://issues.apache.org/jira/browse/HBASE-10772
>             Project: HBase
>          Issue Type: Sub-task
>            Reporter: ramkrishna.s.vasudevan
>            Assignee: ramkrishna.s.vasudevan
>
> Try replacing the BBs with Byte Ranges in Block cache.  See if this can be done in a pluggable way.



--
This message was sent by Atlassian JIRA
(v6.2#6252)