[
https://issues.apache.org/jira/browse/HDFS-1400?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12910836#action_12910836
]
Konstantin Shvachko commented on HDFS-1400:
-------------------------------------------
Agree with Dhruba. Passing Block even in current code would be a good
refactoring.
# I propose to do it in 2 phases. First, refactor trunk by aggregating blockId
and GS into Block parameters across the entire data-transfer code, then
replacing it with ExtendedBlock in the federation branch. This will take a bit
more time now, but will save a lot of time later merging trunk changes to the
branch.
# With your patch you unnecessarily write block length in the serialized
message passed to via DataTransferProtocol. The only thing necessary there is
the requested data length. This is different from current implementation, and
may cause confusion in the future as people wouldn't know that the block length
is actually not required.
> HDFS federation: Introduced block pool ID into DataTransferProtocol
> -------------------------------------------------------------------
>
> Key: HDFS-1400
> URL: https://issues.apache.org/jira/browse/HDFS-1400
> Project: Hadoop HDFS
> Issue Type: New Feature
> Affects Versions: Federation Branch
> Reporter: Suresh Srinivas
> Assignee: Suresh Srinivas
> Fix For: Federation Branch
>
> Attachments: HDFS-1400.patch
>
>
> Block Pool ID needs to be introduced in to DataTransferProtocol
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
