[
https://issues.apache.org/jira/browse/HBASE-10169?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13902000#comment-13902000
]
Gary Helmling commented on HBASE-10169:
---------------------------------------
[[email protected]] Thanks for the comments.
I understand that, for a given request, executing the coprocessor service
invocations in parallel would be beneficial. But I think it would make sense
to add this in a follow on JIRA for a couple of reasons:
# If it's beneficial for coprocessor service invocations, would it also be
beneficial for other batch operations? The answer might be "no" or we might
want to use a separate thread pool for coprocessor service invocations anyway
(so that they don't block / starve other operations). But I think this might
deserve broader discussion.
# I believe it's worth thinking through exactly how we structure the thread
pool given the interaction with RPC handler threads and RPC call queue.
In both cases, I think we can have a more focused discussion and maybe get more
participants by handling the parallelization in a separate JIRA instead of
tacking it on here. Are you okay with that approach?
> Batch coprocessor
> -----------------
>
> Key: HBASE-10169
> URL: https://issues.apache.org/jira/browse/HBASE-10169
> Project: HBase
> Issue Type: Sub-task
> Components: Coprocessors
> Affects Versions: 0.99.0
> Reporter: Jingcheng Du
> Assignee: Jingcheng Du
> Attachments: Batch Coprocessor Design Document.docx,
> HBASE-10169-V2.patch, HBASE-10169-V3.patch, HBASE-10169-V3.patch,
> HBASE-10169-V4.patch, HBASE-10169-V5.patch, HBASE-10169-alternate-2.patch,
> HBASE-10169-alternate-3.patch, HBASE-10169-alternate-4.patch,
> HBASE-10169-alternate.patch, HBASE-10169.patch
>
>
> This is designed to improve the coprocessor invocation in the client side.
> Currently the coprocessor invocation is to send a call to each region. If
> there’s one region server, and 100 regions are located in this server, each
> coprocessor invocation will send 100 calls, each call uses a single thread in
> the client side. The threads will run out soon when the coprocessor
> invocations are heavy.
> In this design, all the calls to the same region server will be grouped into
> one in a single coprocessor invocation. This call will be spread into each
> region in the server side.
--
This message was sent by Atlassian JIRA
(v6.1.5#6160)
