Dear all,
Long time back, I used GMM++ as an interface to BLAS. I have a few questions and i would be thankful if anybody answers:
My application involves dense/space matrix-matrix and matrix-vector multiplication in an iterative scheme. I was wondering if I can involve all cores of a multi-core CPU for this kind of operations. First, I would like to know whether there is any option in GMM++ to activate such property, or I need to link my program to special building of blas/atlas that is multi-threaded.
In case it is not possible, I would be thankful if you suggest a solution perhaps with different library.
Thanks in
advance,
Regards,