[IPython-User] Parallel: Need at least as many tasks as engines?
Wed Nov 10 11:30:22 CST 2010
This is a known issue. We are in the process of completely refactoring the
parallel stuff for 0.11 and I think this one will be fixed automatically in
the course of that refactor. At this point, we do not have plans on fixing
anything in 0.10 ourselves (our time is consumed by 0.11 development), but
we are open to contributions.
On Tue, Nov 9, 2010 at 6:58 AM, John Reid <email@example.com>wrote:
> I'm finding problems when I have more engines than tasks. I'm following
> the example at:
> If I start a cluster with 3 engines thus:
> ipcluster local -n 3
> and then run
> from IPython.kernel import client
> mec = client.MultiEngineClient()
> parallel_result = mec.map(lambda x:x**10, range(2))
> I get an error
> [2:execute]: TypeError: map() requires at least two args
> However if I have just 2 engines running everything is ok. It seems
> strange that the client should have to be aware of the number of engines
> in the cluster. I get a similar problem via the parallel function
> decorator technique.
> I'm using
> Python 2.6.5 (r265:79063, Apr 16 2010, 13:57:41)
> IPython 0.10
> IPython-User mailing list
Brian E. Granger, Ph.D.
Assistant Professor of Physics
Cal Poly State University, San Luis Obispo
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the IPython-User