[
https://issues.apache.org/jira/browse/HBASE-2945?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12904867#action_12904867
]
Jonathan Gray commented on HBASE-2945:
--------------------------------------
The LRU is currently backed by a ConcurrentHashMap and I don't think there's a
way to use byte[] on that, so would require a biggish change.
I thought there was a jira about evicting blocks when the files they came from
got closed but I can't find it. I think it's an interesting direction, just
not sure if it will really be a big win, after all this is what LRUs are for.
Might help with GC by removing references to cached blocks as soon as possible
and en masse.
I think if we are going to do a revamp of the block cache we should come up
with a good set of benchmarks to run. Shouldn't be too bad to run it
standalone and be able to measure the impact of different things as it shows up
in profiling and as it relates to GC and performance.
> stop using string for block name in LRU block cache/hfile
> ---------------------------------------------------------
>
> Key: HBASE-2945
> URL: https://issues.apache.org/jira/browse/HBASE-2945
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.89.20100621
> Reporter: ryan rawson
>
> all my profiling runs indicate there is a large number of string/char[]
> objects and string manipulation is taking a long time in profiling runs.
> These come from the LRU block cache, block id. Also we should support the
> eviction of blocks belonging to particular hfiles, and the current code
> doesn't help with that right now.
> Let's do something better. Whatever that might be.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
