Min Pool Size in WorkflowManager Thread Pool (0.3)

["Resneck, Gabriel M (388J)" <Ga...@jpl.nasa.gov>]

Hi, guys!
I've run into a problem with the Workflow Manager in release 0.3.  The number of active threads allowed by the pool seems to be dictated by the minimum thread count when the queue supplied to the pool object is unlimited.  The default number (6) is a bit low for our purposes, so I was wondering how you guys have dealt with this issue in the past.  Did you simply increase the minimum pool size or implement another solution?  If you increased the minimum pool size, what was the highest that you have used and did you see any issues as a result?
Thanks!

Gabe =)

Re: Min Pool Size in WorkflowManager Thread Pool (0.3)

[Cameron Goodale <si...@gmail.com>]

Gabe,

On Snow I raised the minPoolSize to the number of jobs I am able to run
concurrently on our servers via Resource Manager.

I am running minPoolSize=40 AND maxPoolSize=40 and everything works fine.
 I have noticed that if I flood the Workflow manager too quickly the jobs
tend to wait until all 40 jobs are ready before all of them will move from
STAGING to PGE_EXEC.

Hope that helps.


-Cam


On Mon, Mar 18, 2013 at 10:37 AM, Mattmann, Chris A (388J) <
chris.a.mattmann@jpl.nasa.gov> wrote:

> Hi Gabe,
>
> The best docs I can point you at are here:
>
> http://docs.oracle.com/javase/1.5.0/docs/api/java/util/concurrent/package-s
> ummary.html#package_description
>
>
>
> I would recommend also reading:
>
> http://gee.cs.oswego.edu/dl/classes/EDU/oswego/cs/dl/util/concurrent/Pooled
> Executor.html
>
>
> That really explains what the purpose of those diff properties are.
>
> Cheers,
> Chris
>
>
> On 3/14/13 1:57 PM, "Resneck, Gabriel M (388J)"
> <Ga...@jpl.nasa.gov> wrote:
>
> >Hi, guys!
> >I've run into a problem with the Workflow Manager in release 0.3.  The
> >number of active threads allowed by the pool seems to be dictated by the
> >minimum thread count when the queue supplied to the pool object is
> >unlimited.  The default number (6) is a bit low for our purposes, so I
> >was wondering how you guys have dealt with this issue in the past.  Did
> >you simply increase the minimum pool size or implement another solution?
> >If you increased the minimum pool size, what was the highest that you
> >have used and did you see any issues as a result?
> >Thanks!
> >
> >Gabe =)
> >
>
>


-- 

Sent from a Tin Can attached to a String

Re: Min Pool Size in WorkflowManager Thread Pool (0.3)

["Mattmann, Chris A (388J)" <ch...@jpl.nasa.gov>]

Hi Gabe,

The best docs I can point you at are here:

http://docs.oracle.com/javase/1.5.0/docs/api/java/util/concurrent/package-s
ummary.html#package_description



I would recommend also reading:

http://gee.cs.oswego.edu/dl/classes/EDU/oswego/cs/dl/util/concurrent/Pooled
Executor.html


That really explains what the purpose of those diff properties are.

Cheers,
Chris


On 3/14/13 1:57 PM, "Resneck, Gabriel M (388J)"
<Ga...@jpl.nasa.gov> wrote:

>Hi, guys!
>I've run into a problem with the Workflow Manager in release 0.3.  The
>number of active threads allowed by the pool seems to be dictated by the
>minimum thread count when the queue supplied to the pool object is
>unlimited.  The default number (6) is a bit low for our purposes, so I
>was wondering how you guys have dealt with this issue in the past.  Did
>you simply increase the minimum pool size or implement another solution?
>If you increased the minimum pool size, what was the highest that you
>have used and did you see any issues as a result?
>Thanks!
>
>Gabe =)
>