[
https://issues.apache.org/jira/browse/CASSANDRA-9318?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14531915#comment-14531915
]
Jeremiah Jordan commented on CASSANDRA-9318:
--------------------------------------------
+1 for 2.1. We keep making things faster, in my experience 2.0 and even more so
2.1 are actually easier to make fall over/shed crazy amounts of requests than
earlier versions because we have made the drivers etc so much faster. If we
can do some rate throttling earlier such that things generally slow down
instead of getting a giant spike of timeouts that would be great.
For example when using the spark connector it is very easy to overload a
cluster.
> Bound the number of in-flight requests at the coordinator
> ---------------------------------------------------------
>
> Key: CASSANDRA-9318
> URL: https://issues.apache.org/jira/browse/CASSANDRA-9318
> Project: Cassandra
> Issue Type: Improvement
> Reporter: Ariel Weisberg
> Assignee: Ariel Weisberg
> Fix For: 3.x
>
>
> It's possible to somewhat bound the amount of load accepted into the cluster
> by bounding the number of in-flight requests and request bytes.
> An implementation might do something like track the number of outstanding
> bytes and requests and if it reaches a high watermark disable read on client
> connections until it goes back below some low watermark.
> Need to make sure that disabling read on the client connection won't
> introduce other issues.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
