<div dir="ltr">I think in terms of providing a test for the cluster, the benchmark should not require the shipment of data. For the benchmarking of yt for yt's own sake, then we do need to test the IO performance. <div>
<br></div><div>I think it'd be great to make an in-memory dataset that can then be used for testing performance. For example, set up a bunch of refined spheres and then project/slice/profile/render/halo find? </div>
<div>
<br></div><div style>Sam</div></div><div class="gmail_extra"><br><br><div class="gmail_quote">On Fri, Mar 1, 2013 at 12:31 PM, Matthew Turk <span dir="ltr"><<a href="mailto:matthewturk@gmail.com" target="_blank">matthewturk@gmail.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Hi Jeff,<br>
<br>
Sure. I got asked for a standard yt performance benchmark to be run<br>
to figure out performance of an analysis cluster and evaluate its<br>
readiness. :) I think it's safe to sya that we should be pushing<br>
things like IO, memory capacity and communication performance.<br>
<br>
Adding such a set of scripts (and their results!) would be very useful<br>
going forward, along with a little description of why we chose the<br>
routines we did.<br>
<br>
-Matt<br>
<div class="HOEnZb"><div class="h5"><br>
On Fri, Mar 1, 2013 at 2:17 PM, j s oishi <<a href="mailto:jsoishi@gmail.com">jsoishi@gmail.com</a>> wrote:<br>
> Matt,<br>
><br>
> It might be helpful if you shared some more details of this particular<br>
> benchmarking exercise, if you can. It would be helpful for making sure<br>
> we present the most usefil information to the people asking for it as<br>
> well as info that is useful to us going forwards.<br>
><br>
> j<br>
><br>
> On Fri, Mar 1, 2013 at 2:05 PM, Matthew Turk <<a href="mailto:matthewturk@gmail.com">matthewturk@gmail.com</a>> wrote:<br>
>> Definitely, yes. For what it's worth, the specific use case I have in<br>
>> mind is for performance testing a system, so I think scalability will<br>
>> be important.<br>
>><br>
>> On Fri, Mar 1, 2013 at 2:01 PM, Cameron Hummels <<a href="mailto:chummels@gmail.com">chummels@gmail.com</a>> wrote:<br>
>>> It might be worth showing how well this works on a single proc as well as<br>
>>> how well it works using parallel mode on a few different numbers of<br>
>>> processors.<br>
>>><br>
>>><br>
>>> On Fri, Mar 1, 2013 at 1:57 PM, Christopher Moody <<a href="mailto:chrisemoody@gmail.com">chrisemoody@gmail.com</a>><br>
>>> wrote:<br>
>>>><br>
>>>> Perhaps a benchmark for simply calculating a derived field? This would<br>
>>>> then profile a much smaller codebase.<br>
>>>><br>
>>>> Also benchmarking projections/fields for AMR/octree/particles of similar<br>
>>>> resolutions would be pretty cool.<br>
>>>><br>
>>>> chris<br>
>>>><br>
>>>><br>
>>>> On Fri, Mar 1, 2013 at 10:53 AM, Matthew Turk <<a href="mailto:matthewturk@gmail.com">matthewturk@gmail.com</a>><br>
>>>> wrote:<br>
>>>>><br>
>>>>> Okay, I like that idea. So a unified script with timing for in each<br>
>>>>> section might include:<br>
>>>>><br>
>>>>> * Halo profiling<br>
>>>>> * Global projection<br>
>>>>> * Global profiles<br>
>>>>> * Global VR (CTF and OffAxisProj)<br>
>>>>><br>
>>>>> I think we probably want a large unigrid and a large AMR dataset to<br>
>>>>> run these on, too.<br>
>>>>><br>
>>>>> On Fri, Mar 1, 2013 at 1:50 PM, Nathan Goldbaum <<a href="mailto:goldbaum@ucolick.org">goldbaum@ucolick.org</a>><br>
>>>>> wrote:<br>
>>>>> > Probably a good idea to try some off axis projections or simple volume<br>
>>>>> > renderings to test the VR code.<br>
>>>>> ><br>
>>>>> ><br>
>>>>> > On Fri, Mar 1, 2013 at 10:48 AM, Matthew Turk <<a href="mailto:matthewturk@gmail.com">matthewturk@gmail.com</a>><br>
>>>>> > wrote:<br>
>>>>> >><br>
>>>>> >> Hi all,<br>
>>>>> >><br>
>>>>> >> I got an email asking for a benchmark of yt. I think this is very,<br>
>>>>> >> very valuable to have going forward. I was wondering if anyone had<br>
>>>>> >> any suggestions?<br>
>>>>> >><br>
>>>>> >> I'm thinking that we probably want to test things like the halo<br>
>>>>> >> finder, projections, and phase plots. Would a medium (1536^3) halo<br>
>>>>> >> profiling run do that? Do we want to add on some global projections<br>
>>>>> >> and phase plots as well?<br>
>>>>> >><br>
>>>>> >> -Matt<br>
>>>>> >> _______________________________________________<br>
>>>>> >> yt-dev mailing list<br>
>>>>> >> <a href="mailto:yt-dev@lists.spacepope.org">yt-dev@lists.spacepope.org</a><br>
>>>>> >> <a href="http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org" target="_blank">http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org</a><br>
>>>>> ><br>
>>>>> ><br>
>>>>> ><br>
>>>>> > _______________________________________________<br>
>>>>> > yt-dev mailing list<br>
>>>>> > <a href="mailto:yt-dev@lists.spacepope.org">yt-dev@lists.spacepope.org</a><br>
>>>>> > <a href="http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org" target="_blank">http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org</a><br>
>>>>> ><br>
>>>>> _______________________________________________<br>
>>>>> yt-dev mailing list<br>
>>>>> <a href="mailto:yt-dev@lists.spacepope.org">yt-dev@lists.spacepope.org</a><br>
>>>>> <a href="http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org" target="_blank">http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org</a><br>
>>>><br>
>>>><br>
>>>><br>
>>>> _______________________________________________<br>
>>>> yt-dev mailing list<br>
>>>> <a href="mailto:yt-dev@lists.spacepope.org">yt-dev@lists.spacepope.org</a><br>
>>>> <a href="http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org" target="_blank">http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org</a><br>
>>>><br>
>>><br>
>>><br>
>>><br>
>>> --<br>
>>> Cameron Hummels<br>
>>> Postdoctoral Researcher<br>
>>> Steward Observatory<br>
>>> University of Arizona<br>
>>> <a href="http://chummels.org" target="_blank">http://chummels.org</a><br>
>>><br>
>>> _______________________________________________<br>
>>> yt-dev mailing list<br>
>>> <a href="mailto:yt-dev@lists.spacepope.org">yt-dev@lists.spacepope.org</a><br>
>>> <a href="http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org" target="_blank">http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org</a><br>
>>><br>
>> _______________________________________________<br>
>> yt-dev mailing list<br>
>> <a href="mailto:yt-dev@lists.spacepope.org">yt-dev@lists.spacepope.org</a><br>
>> <a href="http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org" target="_blank">http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org</a><br>
> _______________________________________________<br>
> yt-dev mailing list<br>
> <a href="mailto:yt-dev@lists.spacepope.org">yt-dev@lists.spacepope.org</a><br>
> <a href="http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org" target="_blank">http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org</a><br>
_______________________________________________<br>
yt-dev mailing list<br>
<a href="mailto:yt-dev@lists.spacepope.org">yt-dev@lists.spacepope.org</a><br>
<a href="http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org" target="_blank">http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org</a><br>
</div></div></blockquote></div><br></div>