[Numpy-discussion] Improving Python+MPI import performance

Sturla Molden sturla@molden...
Fri Jan 13 14:38:50 CST 2012


Den 13.01.2012 21:21, skrev Dag Sverre Seljebotn:
> Another idea: Given your diagnostics, wouldn't dumping the output of
> "find" of every path in sys.path to a single text file work well?

It probably would, and would also be less prone to synchronization 
problems than using an MPI broadcast. Another possibility would be to 
use a bsddb (or sqlite?) file as a persistent dict for caching the 
output of imp.find_module.

Sturla



More information about the NumPy-Discussion mailing list