[
https://issues.apache.org/jira/browse/HBASE-7478?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15355397#comment-15355397
]
stack commented on HBASE-7478:
------------------------------
For random reads, I was looking at this and it seemed like we usually respond
inline in the handler and that when we have to service the queue, there is
little in it (sometimes zero entries). This is when there is high throughput.
In general, I'm finding that the less threads we have, the faster we go so
would be interested in evidence of speedup.
> Create a multi-threaded responder
> ---------------------------------
>
> Key: HBASE-7478
> URL: https://issues.apache.org/jira/browse/HBASE-7478
> Project: HBase
> Issue Type: Sub-task
> Reporter: Karthik Ranganathan
>
> Currently, we have multi-threaded readers and handlers, but a single threaded
> responder which is a bottleneck.
> ipc.server.reader.count : number of reader threads to read data off the wire
> ipc.server.handler.count : number of handler threads that process the request
> We need to have the ability to specify a "ipc.server.responder.count" to be
> able to specify the number of responder threads.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
