[ 
https://issues.apache.org/jira/browse/HBASE-15721?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15266979#comment-15266979
 ] 

stack commented on HBASE-15721:
-------------------------------

bq. So instead of allocating a chunk from interface and copy and recreate Cell 
in memstore class, we can allocate and clone within the MSLAB impl. May be we 
need change the comments on class.

I think you are changing MSLAB from being an allocator to instead being a 
manager of the allocations; a change in responsibility. If so, lets work out 
who does what and make roles clear.

Regards Streamable, it does seem like we need methods for 
serializing/deserializing Cells. We can't use 'Writeable' since that type is 
taken (and a mistake we'd like to not repeat). Where do we need Cell 
Serialization and rehydration? From Streams/BBs and byte arrays?

Ok on the getBuffer.

> Optimization in cloning cells into MSLAB
> ----------------------------------------
>
>                 Key: HBASE-15721
>                 URL: https://issues.apache.org/jira/browse/HBASE-15721
>             Project: HBase
>          Issue Type: Sub-task
>          Components: regionserver
>            Reporter: Anoop Sam John
>            Assignee: Anoop Sam John
>             Fix For: 2.0.0
>
>         Attachments: HBASE-15721.patch, HBASE-15721_V2.patch
>
>
> Before cells added to memstore CSLM, there is a clone of cell after copying 
> it to MSLAB chunk area.  This is done not in an efficient way.
> {code}
> public static int appendToByteArray(final Cell cell, final byte[] output, 
> final int offset) {
>     int pos = offset;
>     pos = Bytes.putInt(output, pos, keyLength(cell));
>     pos = Bytes.putInt(output, pos, cell.getValueLength());
>     pos = appendKeyTo(cell, output, pos);
>     pos = CellUtil.copyValueTo(cell, output, pos);
>     if ((cell.getTagsLength() > 0)) {
>       pos = Bytes.putAsShort(output, pos, cell.getTagsLength());
>       pos = CellUtil.copyTagTo(cell, output, pos);
>     }
>     return pos;
>   }
> {code}
> Copied in 9 steps and we end up parsing all lengths.  When the cell 
> implementation is backed by a single byte[] (Like KeyValue) this can be done 
> in single step copy.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Reply via email to