[
https://issues.apache.org/jira/browse/HBASE-5451?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13221122#comment-13221122
]
Devaraj Das commented on HBASE-5451:
------------------------------------
bq. Can we avoid the copy in the interim by having a convention that, if the
request is a protobuf, then we send it following the call envelope rather than
inside it? (does that make sense?)
I explored this route but seems like it's not straightforward to do this (due
to the fact that there are assumptions made on the order of <data-length> and
<data> on the server, and I'd have to make changes to that to accommodate
sending another set of bytes after the call envelope .. messy). I propose we
leave the "copy" around and fix it by introducing something similar to
ProtobufRpcEngine (of Hadoop) that would use native PBs everywhere. Of course
we have to complete moving all protocols to PB.
If people agree with me, I can submit a patch with only the path for the
generated classes changed to what Jimmy suggested.
Thoughts?
> Switch RPC call envelope/headers to PBs
> ---------------------------------------
>
> Key: HBASE-5451
> URL: https://issues.apache.org/jira/browse/HBASE-5451
> Project: HBase
> Issue Type: Sub-task
> Components: ipc, master, migration, regionserver
> Affects Versions: 0.94.0
> Reporter: Todd Lipcon
> Assignee: Devaraj Das
> Fix For: 0.96.0
>
> Attachments: rpc-proto.2.txt, rpc-proto.3.txt, rpc-proto.patch.1_2
>
>
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
