[jira] Updated: (HADOOP-923) DFS Scalability: datanode heartbeat timeouts cause cascading timeouts of other datanodes

["dhruba borthakur (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/HADOOP-923?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

dhruba borthakur updated HADOOP-923:
------------------------------------

    Attachment: scalableheartbeats1.patch

A first version of the patch. Review comments needed.

I compared the two implementation 
Approach 1. keeping a single heartbeat RPC and making the namenode transparently return the list-of-blocks-to-be-transferred once every nth heartbeat call.
Approach 2. Introducing a new RPC call that is invoked by the datanode to retrieve list-of-blocks-to-be-transferred. (default 20 seconds).

This current patch implements Approach 2. This approach allows for having a dedicated namenode thread to process heartbeats (if the need arises).

> DFS Scalability: datanode heartbeat timeouts cause cascading timeouts of other datanodes
> ----------------------------------------------------------------------------------------
>
>                 Key: HADOOP-923
>                 URL: https://issues.apache.org/jira/browse/HADOOP-923
>             Project: Hadoop
>          Issue Type: Bug
>          Components: dfs
>    Affects Versions: 0.10.1
>            Reporter: dhruba borthakur
>         Assigned To: dhruba borthakur
>         Attachments: scalableheartbeats1.patch
>
>
> The datanode sends a heartbeat to the namenode every 3 seconds. The namenode processes the heartbeat and sends  a list of block-to-be-replicated and blocks-to-be-deleted as part of the heartbeat response.
> At times when a couple of datanodes fail, the heartbeat processing on the namenode becomes pretty heavyweight. It acquires the global FSNamesystem lock, traverses the neededReplication structure, generates a list of blocks to be replicated and responds to the heartbeat message. Determining the list of blocks-to-be-replciated is pretty heavyweight, takes plenty of CPU and blocks processing of other heartbeats because of the global FSNamesystem lock.
> It would improve scalability a lot if heartbeat processing does not require the FSNamesystem lock. In fact, the pre-existing "heartbeat" lock already exists for this purpose. 
> I propose that the Heartbeat message be separate from the "retrieve blocks-to-replicate and blocks-to-delete" messages. The datanode can continue to heartbeat once every 3 seconds while it can afford to "retrieve blocks-to-replicate" at a much coarser interval. Heartbeat processing on the namenode will be fast because it does not require the global FSNamesystem lock. Moreover, a datanode failure will not aggrevate the heartbeat processing time on the namenode.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.