hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Anoop Sam John (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-19357) Bucket cache no longer L2 for LRU cache
Date Thu, 07 Dec 2017 17:38:00 GMT

    [ https://issues.apache.org/jira/browse/HBASE-19357?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16282215#comment-16282215
] 

Anoop Sam John commented on HBASE-19357:
----------------------------------------

This is with file BC on HDD only.
bq. This is file-mode but with META data blocks onheap. And it is 20% less QPS compared to
all data blocks being onheap in LRU? 
Yes.  This is with PE randomRead tests.  Ya this is measuring the random read perf for the
user table data.  Not like reads are always against META.
When the META is in file BC, ya  the latency will be more but the reads may be not that frequent
considering the cached location in connection end. But when that happens the latency is more.
bq.File a task for BC file-mode perf eval with particular attention to META data block latencies?
You say in tests read META data blocks specifically? Ya can be done by changing the PE test
code.

> 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)

Mime
View raw message