hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Anoop Sam John (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-16891) Try copying to the Netty ByteBuf directly from the WALEdit
Date Fri, 28 Oct 2016 10:08:58 GMT

    [ https://issues.apache.org/jira/browse/HBASE-16891?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15614990#comment-15614990
] 

Anoop Sam John commented on HBASE-16891:
----------------------------------------

In AsyncFSOutputHelper 
{code}
 @Override
188	      public void writeInt(int i) {
189	        write(Ints.toByteArray(i));
190	      }
191	
192	      @Override
193	      public void write(ByteBuffer bb) {
194	        byte[] dst = new byte[bb.remaining()];
195	        bb.get(dst);
196	        write(dst);
197	      }
{code}
U can just create the 'out' here as org.apache.hadoop.hbase.io.ByteArrayOutputStream object
and then u can just delegate the call to this.out
I mean here
{code}
return new AsyncFSOutput() {

      private final ByteArrayOutputStream out = new ByteArrayOutputStream();
{code}
Else LGTM
Ya we can not avoid the BB duplicate call as netty ByteBuff do not accept any offset pos and
length. It will be bit strange to accept those with a BB on which one can set position and
length.  We did that to avoid garbage. 

> Try copying to the Netty ByteBuf directly from the WALEdit
> ----------------------------------------------------------
>
>                 Key: HBASE-16891
>                 URL: https://issues.apache.org/jira/browse/HBASE-16891
>             Project: HBase
>          Issue Type: Sub-task
>          Components: wal
>    Affects Versions: 2.0.0
>            Reporter: ramkrishna.s.vasudevan
>            Assignee: ramkrishna.s.vasudevan
>             Fix For: 2.0.0
>
>         Attachments: HBASE-16891-v1.patch, HBASE-16891-v2.patch, HBASE-16891.patch
>
>
> -> The FanOutOneBlockAsyncDFSOutput is a much sophisticated dfs client model that
works with Netty ByteBuf. Here we hold on connection to the datanodes using
> Netty Channels. And the idea is to write data direclty to these channels.
> AsyncHLog gets an append call. The AysncWAL uses the HBase's ByteArrayOutputSTream and
so the content of the cell is written to this BAOS and that is again
> copied to the netty Bytebuf in the FanOutOneBlockAsyncDFSOutput.
> So when the sync call happens this FanoutDFSoutput does the checksum calcualtion itself
and then writes the content of this buffer direclty to the DN channel.
> -> In case of FSHLOg this is different. When an append call comes we direclty write
the content to the FSDataOutputStream (it is copied to this stream).
> Then here internally there is a checkSum calculation that happens. when a sync call happens
there is noth ing to do except to notify the NN to flush the latest
> data.
> AS we can see from the above that there are two copies in AsyncWAL
> -> From the Cell to the BAOS 
> -> From the BAOS to the Netty byte buf
> -> On sync() call, do check sum and finally flush the netty byte buf to the DN channel
> In case of FSHLog
> -> From cell to the FSDataoutputstream. data is copied. Check sum happens here.
> -> Sync call just tries to notify the NN.



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

Mime
View raw message