On 09/05/2014 12:08 PM, Ronak Agrawal wrote: > Since the brute force method $c = $a x $b will take 200 hours, can you > guide me with better approach.. >
Multiplication of two matrices is parallelizable using a GPU or multiple CPUs. there are plenty of sample applications that do that natively in C. There is a Perl OpenCL wrapper written as well. You may need to write your matrix multiplication kernel in OpenCL and then use a GPU to do the multiplication. It will get done pretty fast. NVIDIA's Cuda toolkit gives both CUDA and OpenCL sample applications to do Matrix Multiplication. Use them first and see if you can manage it in a small amount of time. --Vikas _______________________________________________ Perldl mailing list Perldl@jach.hawaii.edu http://mailman.jach.hawaii.edu/mailman/listinfo/perldl