[
https://issues.apache.org/jira/browse/HBASE-16417?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15889596#comment-15889596
]
Eshcar Hillel commented on HBASE-16417:
---------------------------------------
I am not running the default values of YCSB.
Value size is set to 100B and number of columns is set to 50. As [~Apache9]
mentioned it could be that the size of value plus size of key is overall >200B,
especially given the number of columns.
bq. And it is not expected to have a small WAL size for AsyncFSWAL.
In NONE (which is the current default in master) I see a reduction in the
overall wal size from 203GB with synchronous WAL to 189GB. Its more than 5%
reduction in WAL size.
But could this be attributed to writing to disk in batches instead of every
entry being synced to disk? Could be that the compression is more effective
with batch writing no?
> In-Memory MemStore Policy for Flattening and Compactions
> --------------------------------------------------------
>
> Key: HBASE-16417
> URL: https://issues.apache.org/jira/browse/HBASE-16417
> Project: HBase
> Issue Type: Sub-task
> Reporter: Anastasia Braginsky
> Assignee: Eshcar Hillel
> Fix For: 2.0.0
>
> Attachments: HBASE-16417-benchmarkresults-20161101.pdf,
> HBASE-16417-benchmarkresults-20161110.pdf,
> HBASE-16417-benchmarkresults-20161123.pdf,
> HBASE-16417-benchmarkresults-20161205.pdf
>
>
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
