[gmx-developers] Performance checklist

[Mark Abraham]

On 13/10/2012 12:34 PM, Shirts, Michael (mrs5pt) wrote:
>> I'd like to set up this sort of testing to be automatic and easy, and
>> have been using Michael's gear for some other work; ideally, these would
>> run via Jenkins.
> I think this would be useful.
>
>> The heavy lifting for 5.0 could break just about
>> anything, and my experience of catching problems with feature X not
>> working with option Y (e.g. md-vv with v-rescale thermostat in 4.5.x,
>> http://redmine.gromacs.org/issues/1000) leads me to the strong opinion
>> that a much broader battery of tests is a very sound (overdue?)
>> investment. They'd need too much time to run for every commit, but
>> catching problems early will be easy and effective with such tests every
>> week or two, or when we know a commit is hitting sensitive areas.
> Roland and I have been talking about this for a while -- we really do need
> something a little more exhaustive to run every week or so.  I assume
> redmine would be a good place to put a plan together?  Any thoughts on this,
> Swedish people who have most of the infrastructure?
>
>> The main caveat is that Michael's checkensemble tools require Python and
>> numpy and to post-process GROMACS output to .xvg. That's fine and good
>> for now, but at some point it could prove worthwhile to get
>> checkensemble to read .edr files.
> I doubt g_energy is the biggest problem timewise, but if it is, then I'd be
> happy to work with getting a .edr reading in.  Only one routine would need
> to be rewritten.  I assume someone someone somewhere has a python reader for
> gromacs .edr files?

I did have a look around recently, but mostly there was a number of stub 
projects that didn't seem to have gone anywhere. :-)

>> Yes, this is the kind of thing I would also like to see, but AFAICR I am
>> yet to hear others' visions for 5.0.
> This is something that would be good to start synthesizing.  Perhaps when
> 4.6 is out the first thing all the developers could do is articulate what
> they want to see for 5.0 so we can synthesize all these ideas?  Would
> Redmine be the best place for this discussion?

People who have perspectives of interest are more likely to find the 
thread here, I think.

Mark