[
https://issues.apache.org/jira/browse/HBASE-19357?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16279930#comment-16279930
]
Anoop Sam John commented on HBASE-19357:
----------------------------------------
bq.Did you find a config that does this?
No there were no extra config. We always used to set META and NS tables to be
with cacheDataInL1. So when there is LRU cache and Bucket cache, we used to put
Meta and NS data in LRU always. Well when the BC is off heap cache, it is
just fine to cache these tables data there. The issue is the file mode.
bq.Hot blocks from a file will be in the os cache
That is true. But we are at the mercy of OS cache then. Having the meta data in
our own on heap cache is always better than expecting OS cache to work for us.
The file mode data cache usage is the key here. When user config this, he knows
abt the pros and cons of this. It is fine. We might not even have to allow user
to config user tables to have cacheDataInL1. This is what Ram says. Ya agree to
that some what. But for META and NS (or system tables we can say) we may not
be able to take such a step?
bq.? We were doing the victim eviction from L1 out to L2 of hbase:meta blocks
also?
The victim eviction was there for each and every block in L1. Not considering
this block belong to what type or which table.
> Bucket cache no longer L2 for LRU cache
> ---------------------------------------
>
> Key: HBASE-19357
> URL: https://issues.apache.org/jira/browse/HBASE-19357
> Project: HBase
> Issue Type: Sub-task
> Reporter: Anoop Sam John
> Assignee: Anoop Sam John
> Fix For: 2.0.0-beta-1
>
> Attachments: HBASE-19357.patch, HBASE-19357.patch,
> HBASE-19357_V2.patch, HBASE-19357_V3.patch, HBASE-19357_V3.patch
>
>
> When Bucket cache is used, by default we dont configure it as an L2 cache
> alone. The default setting is combined mode ON where the data blocks to
> Bucket cache and index/bloom blocks go to LRU cache. But there is a way to
> turn this off and make LRU as L1 and Bucket cache as a victim handler for L1.
> It will be just L2.
> After the off heap read path optimization Bucket cache is no longer slower
> compared to L1. We have test results on data sizes from 12 GB. The Alibaba
> use case was also with 12 GB and they have observed a ~30% QPS improve over
> the LRU cache.
> This issue is to remove the option for combined mode = false. So when Bucket
> cache is in use, data blocks will go to it only and LRU will get only index
> /meta/bloom blocks. Bucket cache will no longer be configured as a victim
> handler for LRU.
> Note : WHen external cache is in use, there only the L1 L2 thing comes. LRU
> will be L1 and external cache act as its L2. That make full sense.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
