charm AT lists.cs.illinois.edu

Subject: Charm++ parallel programming system

List archive

[charm] Timing execution at scale

From: Steve Petruzza <spetruzza AT sci.utah.edu>
To: charm <charm AT lists.cs.illinois.edu>
Subject: [charm] Timing execution at scale
Date: Mon, 8 Aug 2016 12:18:26 +0300

Hi,

In my application I have a global chare array with N tasks, where N can vary from 1K to 1M.

At the moment I am timing the execution from the main chare, where a subset of tasks (10%) at the end will call the “done" function of the main chare.

Do you think that this 10% calls could create a considerable overhead to the timing?

The alternative, I think, could be using a reduction operation on a ProxySection, but creating such a proxy section (it has to be global created by the main chare) with 10% of the tasks plus the reduction operation would not create anyway a bigger overhead (at least in memory)?

If necessary, is there any other alternative to get the timing of a specific part of the execution precisely? Projections could help (at scale)?

Steve

[charm] Timing execution at scale, Steve Petruzza, 08/08/2016
- Message not available
  - Message not available
    - Message not available
      - Message not available
        
        Message not available
        
        Re: [charm] Timing execution at scale, Phil Miller, 08/08/2016
        
        Re: [charm] Timing execution at scale, Steve Petruzza, 08/10/2016
        Re: [charm] Timing execution at scale, Phil Miller, 08/15/2016
        Re: [charm] Timing execution at scale, Steve Petruzza, 08/16/2016
        Re: [charm] Timing execution at scale, Phil Miller, 08/16/2016
        Re: [charm] Timing execution at scale, Steve Petruzza, 08/18/2016