Re: [eigen] Parallelizable operations

[ Thread Index | Date Index | More Archives ]

On 29/07/10 14:22, Matthieu Brucher wrote:

I don't think it would be a huge gain, if there is a gain at all. If
we consider single processor multiple cores, the memory bandwidth is
shared accross cores. Addition and substraction are meory bandwidth
limited, so there would only be additional contention and thus less
performance.Parallelization works best with far more computations than
memory accesses.

Well, there is a simple model that can be evaluated at compile-time w/r to operations cycles count knwown before hand
to decide if parallelization is useful or not.


in NT2 it works OK but then again, we don't have the same compiletime constraints

Mail converted by MHonArc 2.6.19+