Re: Optimisation / parallelisation Problems

[c.]

On 26 Apr 2012, at 11:28, somealias wrote:

> This was exactly what I was looking for! I spent quite a bit of time looking
> at pararrayfun / parcellfun examples but was still unsure how to convert my
> code. I have since applied your code / example for two  functions. I have
> the following performance increases FYI and other interested readers:
> 
> Desktop: Mac Pro 2 CPUs (4 cores each) - total of 16 logical cores
> (hyper-threading)

I think your speed-up would be closer to linear if you don't try to use 
hyperthreading.

> 1. Dataset 1: 810 examples x 6,386 dimensions ~ 8.88x speed up
> 2: Dataset 2: 51,590 examples x 38,449 dimensions ~ 6.44x speed up

I think your best option would be to run both jobs at the same time, each on 4 
cores.

> Thank you so much c.-2! 

c.