[petsc-dev] Subcomms

[Barry Smith]

On Jan 28, 2012, at 2:04 PM, Jed Brown wrote:

> On Sat, Jan 28, 2012 at 13:52, Mark F. Adams <mark.adams at columbia.edu> wrote:
> But I still think its a hard problem.  There are many constraints on a language and the inertia for science applications is immense.  The risk of failure is more than most application scientist can tolerate.  (eg, I recently heard of a code that decided to use python, which does not sound risky to me, but it takes 2 hours for python to load on 64K cores.  There are many failure modes in this business)
> 
> I agree that it is hard. FWIW, Aron Ahmadia and I worked out the key to fixing the dynamic loading issue. We're working out how to deploy it now.

   Come on, getting python to load properly on a big system is not hard (after all you and Aron figured out how to do it). It is just that those with any power to change the situation simply don't give a fucking shit about this kind of practical issue. How many times have people running the "supercomputing centers" ever made an effort to improve things for users? Not often, they just like to bask in their power and incompetence. And yes, this has nothing to do with intelligence, just self-interest.

   Barry