As discussed here earlier this year, the DARPA HPCS program has had a lot of positive publicity about its focus on usability and productivity in HPC systems. The question of how to evaluate HPC technologies in terms of productivity has been the subject of quite a bit of research funded under the first two phases of the HPCS program. Take a look at the highproductivity.org publications list for some of the results.
Recently, it has come to light that funding for productivity evaluation research under HPCS is going away after the end of Phase 2. We asked for public comment from DARPA on this subject, and here is what we got back from Jan Walker, the DARPA spokesman:
insideHPC: Is it true that a change was made in funding priorities between phase 2 & 3 of HPCS that resulted in no funding for university researchers in phase 3?
DARPA: Phase III is not a phase that requires as much research as past phases – it is focused on development and building of a prototype. However, if Phase III contractors IBM and Cray want to fund researchers they certainly can.
insideHPC: Will the productivity evaluation research activities continue? Who will be doing it, and if not, why not?
DARPA: Productivity evaluations on the HPCS systems will continue. Some will be done by the vendors themselves. These evaluations will be a self-evaluation to make sure they are on track for productivity gains.
One interpretation of this is that since the Phase 3 vendors IBM and Cray will be allowed to evaluate themselves, it would be a big surprise if they failed to reach the productivity goals in the contract.