[
https://issues.apache.org/jira/browse/HADOOP-3412?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Tom White updated HADOOP-3412:
------------------------------
Attachment: JobScheduler-v9.patch
Here is a new patch which takes Brice's good work and strips it back to
minimise the changes to JobTracker to provide a start for future work on
scheduling, such as HADOOP-3445. In a nutshell, the major change here is the
removal of the getNewTaskForTaskTracker method from JobTracker, replacing it
with a call to TaskScheduler#assignTask, to meet the goal of supporting
pluggable schedulers.
TaskScheduler has the three methods Matei mentioned: addJob, removeJob,
assignTask. It also has a reference to the JobTracker, via a TaskTrackerManager
interface, which is like TaskTrackerContainer in the last patch and which I
introduced to permit testing. (I've put a comment in it to say as much.)
I've renamed DefaultTaskScheduler to JobQueueTaskScheduler to better describe
what it does, and I've replaced the JobQueue abstraction with a
SortedSet<JobInProgress>. My feeling is that a SortedSet is sufficient to do
job prioritization, since you can specify a Comparator to impose an order on
the queue. More complex schedulers can just implement the three TaskScheduler
methods, which shouldn't be too much of a burden. This is an area where I
imagine we can make improvements after this change has been committed.
Finally, I've enhanced the unit test to test several scheduling scenarios -
this test directly tests the scheduling algorithm by providing a (static) mock
TaskTrackerManager implementation.
> Refactor the scheduler out of the JobTracker
> --------------------------------------------
>
> Key: HADOOP-3412
> URL: https://issues.apache.org/jira/browse/HADOOP-3412
> Project: Hadoop Core
> Issue Type: Improvement
> Components: mapred
> Reporter: Brice Arnould
> Assignee: Brice Arnould
> Priority: Minor
> Fix For: 0.19.0
>
> Attachments: JobScheduler-v9.patch, JobScheduler.patch,
> JobScheduler_v2.patch, JobScheduler_v3.patch, JobScheduler_v3b.patch,
> JobScheduler_v4.patch, JobScheduler_v5.patch, JobScheduler_v6.1.patch,
> JobScheduler_v6.2.patch, JobScheduler_v6.3.patch, JobScheduler_v6.4.patch,
> JobScheduler_v6.patch, JobScheduler_v7.1.patch, JobScheduler_v7.patch,
> JobScheduler_v8.patch, RackAwareJobScheduler.java,
> SimpleResourceAwareJobScheduler.java
>
>
> First I would like warn you that my proposition is assumed to be very naive.
> I just hope that reading it won't make you lose time.
> h4. The aim
> It seems to me that improving Hadoop scheduling could be very profitable.
> But, it is hard to implement and compare schedulers, because the scheduling
> logic is mixed within the rest of the JobTracker.
> This bug is the first step of an attempt to improve the Hadoop scheduler. It
> re-implements the current scheduling algorithm in a separate class called
> JobScheduler. This new class is instantiated in the JobTracker.
> h4. Bug fixed as a side effects
> This patch probably cannot be submited as it is.
> A first difficulty is that it does not have exactly the same behaviour than
> the current JobTracker. More precisely, it doesn't re-implement things like
> code that seems to be never called or concurency problems.
> I wrote TOCONFIRM where my proposition differ from the current
> implementation, so you can find them easily.
> I know that fixing bugs silently is bad. So, independently of what you decide
> about this patch, I will open issues for bugs that you confirm.
> h4. Other side effects
> Another side effect of this patch is to add documentation about each step of
> the scheduling. I hope that it will help future improvement by lowering the
> level required to contribute to the scheduler.
> It also reduces the complexity and the granularity of the JobTracker (making
> it more parallel).
> h4. The future
> If you feel that this is a step the right direction, I will try to propose a
> JobSchedulerInterface that many JobSchedulers could implement and to propose
> alternatives to the current « FifoJobScheduler ». If some of you have ideas
> about that please tell ^^ I will also open issues for things marked as FIXME
> in the patch.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
