[
https://issues.apache.org/jira/browse/HADOOP-7557?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13199036#comment-13199036
]
Benoit Sigoure commented on HADOOP-7557:
----------------------------------------
If I can just throw in my 2¢: I implemented an HBase client entirely from
scratch, so I had to figure out the Hadoop RPC protocol and its HBase variant –
which BTW I [documented
here|https://github.com/stumbleupon/asynchbase/blob/master/src/HBaseRpc.java#L135].
I completely agree with what Todd said. I wouldn't have said it better myself,
so let me quote him:
bq. To be honest, all of this pluggability in serializations seems to be
counter to the goal of interoperability here. If we just use protobufs
everywhere, than anyone can come along in any reasonably common language and
get a working implementation of Hadoop RPC on either the server or client side.
If every single piece has a flag indicating whether it might be protobuf, avro,
Writable, msgpack, JSON, etc, then anyone trying to implement the protocol will
be faced with an insurmountable amount of work.
The current Hadoop and HBase RPC protocols are unnecessarily hard to implement.
Using an unified serialization mechanism such as PB for everything is a great
step towards making clients significantly simpler.
Simpler is better. KISS.
> Make IPC header be extensible
> -------------------------------
>
> Key: HADOOP-7557
> URL: https://issues.apache.org/jira/browse/HADOOP-7557
> Project: Hadoop Common
> Issue Type: Sub-task
> Reporter: Sanjay Radia
> Assignee: Sanjay Radia
> Attachments: HADOOP-7557.patch, IpcHeader.proto, ipcHeader1.patch,
> ipcHeader2.patch
>
>
--
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
