[
https://issues.apache.org/jira/browse/CASSANDRA-18120?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17866799#comment-17866799
]
Dan Sarisky commented on CASSANDRA-18120:
-----------------------------------------
[~shunsaker] This looks great. Both dynamic and dynamic_remote did what we
wanted. When a single node is crippled, there is no longer a 90% reduction in
throughput. It is much closer to 1/number_over_nodes reduction, which is what
one logically expects.
Thanks!
> Single slow node dramatically reduces cluster logged batch write throughput
> regardless of CL
> --------------------------------------------------------------------------------------------
>
> Key: CASSANDRA-18120
> URL: https://issues.apache.org/jira/browse/CASSANDRA-18120
> Project: Cassandra
> Issue Type: Improvement
> Components: Consistency/Coordination
> Reporter: Dan Sarisky
> Assignee: Maxim Chanturiay
> Priority: Normal
> Fix For: 4.0.x, 4.1.x, 5.0.x, 5.x
>
>
> We issue writes to Cassandra as logged batches(RF=3, Consistency levels=TWO,
> QUORUM, or LOCAL_QUORUM)
>
> On clusters of any size - a single extremely slow node causes a ~90% loss of
> cluster-wide throughput using batched writes. We can replicate this in the
> lab via CPU or disk throttling. I observe this in 3.11, 4.0, and 4.1.
>
> It appears the mechanism in play is:
> Those logged batches are immediately written to two replica nodes and the
> actual mutations aren't processed until those two nodes acknowledge the batch
> statements. Those replica nodes are selected randomly from all nodes in the
> local data center currently up in gossip. If a single node is slow, but
> still thought to be up in gossip, this eventually causes every other node to
> have all of its MutationStages to be waiting while the slow replica accepts
> batch writes.
>
> The code in play appears to be:
> See
> [https://github.com/apache/cassandra/blob/trunk/src/java/org/apache/cassandra/locator/ReplicaPlans.java#L245].
> In the method filterBatchlogEndpoints() there is a
> Collections.shuffle() to order the endpoints and a
> FailureDetector.isEndpointAlive() to test if the endpoint is acceptable.
>
> This behavior causes Cassandra to move from a multi-node fault tolerant
> system toa collection of single points of failure.
>
> We try to take administrator actions to kill off the extremely slow nodes,
> but it would be great to have some notion of "what node is a bad choice" when
> writing log batches to replica nodes.
>
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
