Python element wise operations in parallel

Asked Apr 20 '17 at 16:13

Active Apr 20 '17 at 16:48

Viewed 239 times

Is there a simple way to do element wise numpy array operations of python in parallel.

I am trying to do a = b * c + d (all 3d arrays, [9, 200, 200] ). I can't get this to run in parallel because htop (on Linux) shows me only one thread being used when I run the above step for thousands of iterations. I ran it on a cluster too with multiple nodes, but same results.

Tried both Python 2 and 3.

No independent for-loops, so cython won't help much?

All I need is just simple addition/multiplication of numpy arrays in parallel.

Thanks

edited Apr 20 '17 at 16:48

kennytm

510,854
105
1,084
1,005

asked Apr 20 '17 at 16:13

RaVi

http://cython.readthedocs.io/en/latest/src/userguide/parallelism.html – user2357112 Apr 20 '17 at 16:46
1

The linked duplicate is several years old. There may be some newer discussions that are just as good. – hpaulj Apr 20 '17 at 17:14

Python element wise operations in parallel

0 Answers0