Duo Zhang commented on HBASE-15536:

Oh only one DN? I haven't tested this scenario before. In general, AsyncFSWAL 
will have a lower latency if you write to more than 1 DN as it sends the data 
to multiple DNs concurrently.

And it is interesting that WALPE and PE show different result. Will check the 
checksum work you mentioned.

And I think we need to do some profiling work to see where is the hot spot.


> Make AsyncFSWAL as our default WAL
> ----------------------------------
>                 Key: HBASE-15536
>                 URL: https://issues.apache.org/jira/browse/HBASE-15536
>             Project: HBase
>          Issue Type: Sub-task
>          Components: wal
>    Affects Versions: 2.0.0
>            Reporter: Duo Zhang
>            Assignee: Duo Zhang
>            Priority: Critical
>             Fix For: 2.0.0
>         Attachments: HBASE-15536-v1.patch, HBASE-15536-v2.patch, 
> HBASE-15536-v3.patch, HBASE-15536-v4.patch, HBASE-15536-v5.patch, 
> HBASE-15536.patch
> As it should be predicated on passing basic cluster ITBLL

This message was sent by Atlassian JIRA

Reply via email to