On Thu, Jan 03, 2013 at 01:11:40PM -0800, Iuri G. wrote:
>
> Manny thanks for your outstanding support of ruote.

Hello Iuri,

thanks and happy new year!

> I have been thinking to set up two workers for my processes where one
> worker will only deal with short running processes and another will deal
> only with long running ones.
> I was wondering if you have any recommendation for this kind of setup.

If I take your sentence above literally with "worker" == "ruote worker". Then
may I suggest using one engine/dashboard for short processes and another for
long processes? That would completely separate things.


> Only implementation I can come up with is to use two storage
> participants, but that would complicated code a lot.

Having short processes use participant "a" and long processes use participant
"b" sounds not too complicated.


> I read about storage participant method #accept?, but that gives affinity
> to storage participant and not to worker that processes it.

A storage participant is very simple, it just receives the workitem and
stores it, there is isn't much work going on. I don't think having dedicated
workers per workflow is bringing anything.

I'm not sure I understand your question. Maybe you can rephrase it and we can
discuss further...


Cheers,

--
John Mettraux - http://lambda.io/jmettraux

-- 
you received this message because you are subscribed to the "ruote users" group.
to post : send email to [email protected]
to unsubscribe : send email to [email protected]
more options : http://groups.google.com/group/openwferu-users?hl=en

Reply via email to