[Yt-dev] yt release discussion

[Matthew Turk]

Hi all,

On Thu, Nov 10, 2011 at 10:45 AM, Stephen Skory <s at skory.us> wrote:
> All,
>
>> https://bitbucket.org/yt_analysis/yt/issues?status=new&status=open&sort=milestone&milestone=2.3
>
> I'm going to see if I can knock off my task from this list above.
> Namely, this one
>
> https://bitbucket.org/yt_analysis/yt/issue/247/estimates-of-resources-for-parallelism

Thanks, Stephen.  Your outline sounds good.

I just killed off #287:

https://bitbucket.org/yt_analysis/yt/issue/287/leaflet-gmaps-interface-in-reason

I think a bunch of the open bugs can be pushed back to a later time,
but I think it would be best if issue owners did so themselves.

>
> I think instead of doing a buch of very time-consuming tests and
> benchmarks, I'm going to just write up a page that specifically
> describes the kinds of parallelism in yt, the sources of CPU time,
> memory usage, and disk access, and perhaps some insight in the optimal
> numbers of processors and memory. I'll do most of the writing, but I
> may call on a few of you to help me fill in the gaps in my knowledge
> (volume rendering, light rays/cones, for example). Here's what I've
> got.
>
> domain-based parallelism: halo finding, two point functions
>
> object-based parallelism: halo profiler, clump finding,
>
> grid-based parallelism: projections, quantities, (I'll do a search for
> _get_grid_objs())
>
> I know I'm missing a bunch of stuff, so could you take a second to
> fill in what I can't think of right now? For example, I'm not sure
> what the best classification is for volume rendering. Thanks!
>

I think VR right now is not grid-based, although my understanding is
that we have a plan for moving it in that direction ... Sam can speak
more to that.

I begrudgingly accept Sam's new release codename.  But: this isn't over!

-Matt

> --
> Stephen Skory
> s at skory.us
> http://stephenskory.com/
> 510.621.3687 (google voice)
> _______________________________________________
> Yt-dev mailing list
> Yt-dev at lists.spacepope.org
> http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
>