[yt-users] timing parallel runs

Matthew Turk matthewturk at gmail.com
Tue Mar 8 17:46:41 PST 2011


Hi Dave,

Stephen's on the mark; not everything works yet, but it's actively
being developed, and that page mostly gets you up to date.  We're also
working to interface with Enzo's existing inline halo finder still.

Britton Smith and I are taking the lead on this project, and we're
aiming to have some results -- good or bad -- by the end of March.
This project, and ensuring its scalability, is actually a driver and
necessary components for some of the next generation simulations I
will be conducting and so I'm very keen and motivated to do this.
Furthermore, it's an aspect of my current NSF fellowship.

Right now it works quite nicely for small runs, up to about 64
processors.  When we have better and more concrete results by the end
of March or so, I will communicate them here on the list, in the yt
docs, and I will also reach out to you personally.

Best,

Matt

On Tue, Mar 8, 2011 at 12:26 PM, Stephen Skory <stephenskory at yahoo.com> wrote:
> Hi Dave,
>
>> Now that I think of it timing slices man not be the ideal thing to test. I will
>> have a look at your pointers. How is the in-situ work on enzo going. Is there a
>> section in the YT docs that deals with using YT in this fashion?
>
>
> I have some inside knowledge that Matt is mostly incommunicado for the next couple days, so I'll field your question.
>
> You should look at the enzo documentation, here, for some information on inline analysis.
>
> http://docs.enzo.googlecode.com/hg/user_guide/EmbeddedPython.html
>
>
> Not everything works inline in parallel yet, and I've seen a list of what does somewhere, but I can't locate it right now. Off the top of my head I know that halo finding (using Parallel HOP), non-adaptive projections, and profiles work inline. I think there might be a couple more things...
>
> Good luck!
>
>
> Stephen Skory
> stephenskory at yahoo.com
> http://stephenskory.com/
> 510.621.3687 (google voice)
>
> _______________________________________________
> yt-users mailing list
> yt-users at lists.spacepope.org
> http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
>



More information about the yt-users mailing list