[
https://issues.apache.org/jira/browse/HBASE-15765?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15285893#comment-15285893
]
Heng Chen commented on HBASE-15765:
-----------------------------------
{quote}
we can steal 1 RPC every ~3seconds if I read the patch right.
{quote}
It is not steal task every 3 seconds. In the patch, we firstly get task from
our original queue, if timeout (3s), then we will try to steal task from other
queue.
{quote}
the handlers still do not steal from the queues in the same tier, but a
different "shard"
{quote}
Sorry, i am not sure what is "queues in the same tier"?
If i understand correctly, the queues in one handler pools are in the same
priority level, right? Or you mean general pool and replication pool are in
the same tier?
> RPC handlers should steal from other queues in the same or higher priority
> --------------------------------------------------------------------------
>
> Key: HBASE-15765
> URL: https://issues.apache.org/jira/browse/HBASE-15765
> Project: HBase
> Issue Type: Improvement
> Reporter: Enis Soztutar
> Attachments: HBASE-15765.patch, HBASE-15765_v1.patch
>
>
> With the simple rpc scheduler model by default, a request first falls in one
> type of queue (priority, general, replication) and it will get "sharded" in a
> queue within that priority level (for example there are 2 queues for the high
> priority level). The handlers can only take from their specific queues, and
> nothing else.
> Even if a request is waiting in a high priority queue, the handlers within
> the same priority level or a lower priority level will not see these requests
> (which is the root cause for HBASE-15613). We should be able to steal from
> other queues, without losing the prioritization and sharding benefits.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
