[
https://issues.apache.org/jira/browse/HADOOP-1985?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Devaraj Das updated HADOOP-1985:
--------------------------------
Attachment: jobinprogress.patch
Here is a patch with some changes in the task cache datastructure. Also, there
are changes that tries to ensure that rack (and higher level) locality is
preserved for failed/speculative tasks also... It doesn't delete TIPs from a
node cache until the TIP is complete or the node happens to be the host itself.
The logic is that we should not delete TIPs from the rack level cache to avoid
having the speculative/failed TIPs pay a peformance penalty if some other
tasktracker from the same rack gets to run that failed/speculative task. We
delete TIPs from the host cache since we don't execute the same tip on a host
that failed to execute it earlier...
Would highly appreciate a quick review on this one.
> Abstract node to switch mapping into a topology service class used by
> namenode and jobtracker
> ---------------------------------------------------------------------------------------------
>
> Key: HADOOP-1985
> URL: https://issues.apache.org/jira/browse/HADOOP-1985
> Project: Hadoop
> Issue Type: New Feature
> Components: dfs, mapred
> Reporter: eric baldeschwieler
> Assignee: Devaraj Das
> Fix For: 0.16.0
>
> Attachments: 1985.new.patch, 1985.v1.patch, 1985.v2.patch,
> 1985.v3.patch, 1985.v4.patch, 1985.v5.patch, 1985.v6.patch,
> jobinprogress.patch
>
>
> In order to implement switch locality in MapReduce, we need to have switch
> location in both the namenode and job tracker. Currently the namenode asks
> the data nodes for this info and they run a local script to answer this
> question. In our environment and others that I know of there is no reason to
> push this to each node. It is easier to maintain a centralized script that
> maps node DNS names to switch strings.
> I propose that we build a new class that caches known DNS name to switch
> mappings and invokes a loadable class or a configurable system call to
> resolve unknown DNS to switch mappings. We can then add this to the namenode
> to support the current block to switch mapping needs and simplify the data
> nodes. We can also add this same callout to the job tracker and then
> implement rack locality logic there without needing to chane the filesystem
> API or the split planning API.
> Not only is this the least intrusive path to building racklocal MR I can ID,
> it is also future compatible to future infrastructures that may derive
> topology on the fly, etc, etc...
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
