[
https://issues.apache.org/jira/browse/HBASE-13170?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Elliott Clark updated HBASE-13170:
----------------------------------
Attachment: HBASE-13170.patch
Here's the latest version.
Decent configs.
Better stats handling
better handling of double checked locking for block cache ( this was a huge per
win)
I'd propose that we get something like this patch in, and then re-do config and
multi-tiered caching in another jira. I think that re-do will have to be a
breaking change, and it would be nice to decouple the feature from the breaking
change.
To test I have been running ltt and when I kill the local instance and let it
come back, I have > 60% cache hit ratio in 10 seconds after opening the
regions. Overall response times are the same.
> Allow block cache to be external
> --------------------------------
>
> Key: HBASE-13170
> URL: https://issues.apache.org/jira/browse/HBASE-13170
> Project: HBase
> Issue Type: New Feature
> Components: io
> Affects Versions: 2.0.0
> Reporter: Elliott Clark
> Assignee: Elliott Clark
> Fix For: 1.1.0
>
> Attachments: HBASE-13170.patch
>
>
> Allow an external service to provide the block cache. This has the nice
> property of allowing failover/upgrades to happen without causing a fully cold
> cache.
> Additionally this allows read replicas to share some of the same memory.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
