[
https://issues.apache.org/jira/browse/HBASE-12790?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14996945#comment-14996945
]
Andrew Purtell commented on HBASE-12790:
----------------------------------------
If we can all conceptually agree on round robin scheduling of connections then
the gap here is pretty small.
bq. Opening a new HConnection takes three seconds. It's not feasible to pool
enough HConnections to support the concurrency we see over the various types of
queries.
Then can we make this faster?
For example, when creating a new Connection object, it is not strictly
necessary this must map 1:1 with communication channels at the RPC layer or TCP
connections on the wire. Why can't a new Connection transparently re-use the
communication channel constructed by a previous instance? In which case the
distinction between the new Connection and previous is a logical identifier. A
logical identifier carried in requests that can be used to schedule round robin
on the server. A logical identifier which does not need to be exposed to the
application. The application has control of round-robin dispatch through its
allocation of Connection objects.
> Support fairness across parallelized scans
> ------------------------------------------
>
> Key: HBASE-12790
> URL: https://issues.apache.org/jira/browse/HBASE-12790
> Project: HBase
> Issue Type: New Feature
> Reporter: James Taylor
> Assignee: ramkrishna.s.vasudevan
> Labels: Phoenix
> Attachments: AbstractRoundRobinQueue.java, HBASE-12790.patch,
> HBASE-12790_1.patch, HBASE-12790_5.patch, HBASE-12790_callwrapper.patch,
> HBASE-12790_trunk_1.patch, PHOENIX_4.5.3-HBase-0.98-2317-SNAPSHOT.zip
>
>
> Some HBase clients parallelize the execution of a scan to reduce latency in
> getting back results. This can lead to starvation with a loaded cluster and
> interleaved scans, since the RPC queue will be ordered and processed on a
> FIFO basis. For example, if there are two clients, A & B that submit largish
> scans at the same time. Say each scan is broken down into 100 scans by the
> client (broken down into equal depth chunks along the row key), and the 100
> scans of client A are queued first, followed immediately by the 100 scans of
> client B. In this case, client B will be starved out of getting any results
> back until the scans for client A complete.
> One solution to this is to use the attached AbstractRoundRobinQueue instead
> of the standard FIFO queue. The queue to be used could be (maybe it already
> is) configurable based on a new config parameter. Using this queue would
> require the client to have the same identifier for all of the 100 parallel
> scans that represent a single logical scan from the clients point of view.
> With this information, the round robin queue would pick off a task from the
> queue in a round robin fashion (instead of a strictly FIFO manner) to prevent
> starvation over interleaved parallelized scans.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
