[
https://issues.apache.org/jira/browse/CASSANDRA-9738?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14741403#comment-14741403
]
Ariel Weisberg commented on CASSANDRA-9738:
-------------------------------------------
bq. Is that the entire partition's worth of index then that we're copying on
heap?
Yes
bq. Can we use the partition size statistics to automatically pick the
appropriate implementation?
Possibly. If we have two cache implementations at once how do you configure how
big each one is? It seems tricky to do in one week.
I think we would have better luck trying to get reference counting working so
there doesn't need to be a copy. In a lot of cases the RIE is going to be
enclosed by a Closable iterator anyways.
> Migrate key-cache to be fully off-heap
> --------------------------------------
>
> Key: CASSANDRA-9738
> URL: https://issues.apache.org/jira/browse/CASSANDRA-9738
> Project: Cassandra
> Issue Type: Sub-task
> Reporter: Robert Stupp
> Assignee: Robert Stupp
> Fix For: 3.0.0 rc1
>
>
> Key cache still uses a concurrent map on-heap. This could go to off-heap and
> feels doable now after CASSANDRA-8099.
> Evaluation should be done in advance based on a POC to prove that pure
> off-heap counter cache buys a performance and/or gc-pressure improvement.
> In theory, elimination of on-heap management of the map should buy us some
> benefit.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
