[ https://issues.apache.org/jira/browse/CASSANDRA-9318?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14531066#comment-14531066 ]
Jonathan Ellis commented on CASSANDRA-9318: ------------------------------------------- There's a lot of overlap, but bounding request count + size is definitely simpler than trying to bound the heap used more generally. I think the former is simple enough, and the problem it solves serious enough (nodes falling over OOM and/or dropping lots of requests in attempt to load shed) that we should do it in 2.1 if possible. This is a big pain point for C* users in the real world. > 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)