[
https://issues.apache.org/jira/browse/HBASE-12848?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14619147#comment-14619147
]
Andrew Purtell commented on HBASE-12848:
----------------------------------------
bq. HDFS wont do the movement of block data across diff devices when the rename
happens. There is a mover CLI tool for doing so which has to be called
explicitly.
Manual CLI tool? That's a bit lame.
I made a proposal once for a scheme where the NN would store storage policy for
various paths as xattrs, provide device class hints to the datanode during
block allocation and replication according to defined policy, and automatically
coordinate block movement from one storage tier to another should the storage
policy xattr be changed for a given path. Not worth reviving that particular
JIRA, but much of the groundwork for this is now in place in HDFS. Maybe
someone could propose the remaining missing pieces?
> Utilize Flash storage for WAL
> -----------------------------
>
> Key: HBASE-12848
> URL: https://issues.apache.org/jira/browse/HBASE-12848
> Project: HBase
> Issue Type: Sub-task
> Reporter: Ted Yu
> Assignee: Ted Yu
> Fix For: 2.0.0, 1.1.0
>
> Attachments: 12848-v1.patch, 12848-v2.patch, 12848-v3.patch,
> 12848-v4.patch, 12848-v4.patch
>
>
> One way to improve data ingestion rate is to make use of Flash storage.
> HDFS is doing the heavy lifting - see HDFS-7228.
> We assume an environment where:
> 1. Some servers have a mix of flash, e.g. 2 flash drives and 4 traditional
> drives.
> 2. Some servers have all traditional storage.
> 3. RegionServers are deployed on both profiles within one HBase cluster.
> This JIRA allows WAL to be managed on flash in a mixed-profile environment.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
