Hi Akio
I think that the theoretical ratio is much bigger than what you would be able to achieve in practice. It's hard to compute exactly, but for example for em_reg it should be the number of samples (which are processed independently), so something on the order of 1000. For ca_reg it should be bigger, but it's a bit more complicated as things aren't independent.
cheers Bruce
On Tue, 3 Jul 2012, Akio Yamamoto wrote:
Freesurfer experts,
I have been working on openmp parallelization of Freesurfer programs.
I'd like to know the ratio of sequential processing parts (which are essentially impossible to be parallelized) to the total amount of processing in order to calculate the theoretical limit of performance improvement rate.
Have you ever made an estimate of the ratio between sequential and parallel processing segments, especially for ca_reg and em_reg?
Thanks, Akio
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer