[
https://issues.apache.org/jira/browse/HDFS-1400?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12909450#action_12909450
]
Suresh Srinivas commented on HDFS-1400:
---------------------------------------
Currently DataTransferProtocol has methods such as:
{noformat}
public static void opReadBlock(DataOutputStream out, long blockId,
long blockGs, long blockOffset, long blockLen, String clientName,
Token<BlockTokenIdentifier> blockToken) throws IOException;
{noformat}
The client has to pass the individual elements that make block identification
such as blockId and generation stamp. While adding block pool ID, I propose
methods with the following format:
{noformat}
public static void opReadBlock(DataOutputStream out, ExtendedBlock block,
long blockOffset, long blockLen, String clientName,
Token<BlockTokenIdentifier> blockToken) throws IOException;
{noformat}
With this, the client need not understand the internals of ExtendedBlock. It
receives ExtendedBlock over RPC and sends it in DataTransferProtocol. This
helps in making ExtendedBlock opaque to the client.
> 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
>
>
> 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.
