[
https://issues.apache.org/jira/browse/MAPREDUCE-4495?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13427380#comment-13427380
]
Alejandro Abdelnur commented on MAPREDUCE-4495:
-----------------------------------------------
@Chris, as we all know, doing a few JIRAs to add new functionality to Hadoop is
completely different from bootstrapping an incubator project. We can easily
avoid circular dependencies, thus we should. Regarding you 3 questions, only
time will tell. I'm not opposed to a separate project, I'm just worried about
starting a new project without a need for it, and only time and adoption will
tell (I'm trying not to go 'project happy' here). At that point we can split it
from Hadoop as it has been done with other projects.
@Arun, for users of the DAG-AM, the NodeHandler interface is the equivalent to
the Mapper/Reducer interfaces in MapReduce. The proposed Workflow library has a
NodeHandler interface which is called when the workflow job enters a node and
when it exits the node. You can define/user as many NodeHandler implementations
you need in a workflow. If DAG-AM lives in mapreduce, it would provide out the
box a NodeHandler implementation for handle MR jobs. Then, if Pig/Hive/Oozie
use the DAG-AM they could use the built-in MR NodeHandler for MR jobs and
provide their own NodeHandlers for other types of tasks. I've made the analogy
with the JobControl not with the intention of providing a replacement for it
but to explain the functionality you'd get with the DAG-AM. Hope this clarifies.
> Workflow Application Master in YARN
> -----------------------------------
>
> Key: MAPREDUCE-4495
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-4495
> Project: Hadoop Map/Reduce
> Issue Type: New Feature
> Affects Versions: 2.0.0-alpha
> Reporter: Bo Wang
> Assignee: Bo Wang
>
> It is useful to have a workflow application master, which will be capable of
> running a DAG of jobs. The workflow client submits a DAG request to the AM
> and then the AM will manage the life cycle of this application in terms of
> requesting the needed resources from the RM, and starting, monitoring and
> retrying the application's individual tasks.
> Compared to running Oozie with the current MapReduce Application Master,
> these are some of the advantages:
> - Less number of consumed resources, since only one application master will
> be spawned for the whole workflow.
> - Reuse of resources, since the same resources can be used by multiple
> consecutive jobs in the workflow (no need to request/wait for resources for
> every individual job from the central RM).
> - More optimization opportunities in terms of collective resource requests.
> - Optimization opportunities in terms of rewriting and composing jobs in the
> workflow (e.g. pushing down Mappers).
> - This Application Master can be reused/extended by higher systems like Pig
> and hive to provide an optimized way of running their workflows.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
