On 09/26/2011 10:57 AM, Tarek Elsayed wrote:

I tried Eigen and MTL4. Neither could compete with ublas in its slowness ! If you think of going to GMM++, you can also try MTL4, I found it very fast in sparse matrix computations. If you want more speed, then you can use the intel MKL library.

Dear Tarek,

Thanks for the reply,
I tried to use Intel MKL for sparse matrix-dense vector multiplication and found out that gmm++ was faster, I used g++ for these tests.

To be honest, I was surprised that gmm++ could beat intel MKL but did not have time to make further tests. Do you have some simple code that I might have a look at for interfacing MKL and MTL4. And one more question, did you use intel compiler for there tests?

I have tried both MKL and MTL4 simultaneously. MTL4 can provide you with the pointers to pass for MKL functions, so you don't have to rewrite your entire code.

Oh, this is a good pointer and remark.

Since each of MTL4, eigen, gmm++ are tempelated libraries like ublas, I think you don't have to change much of your code to benchmark each of them.

es selamu alejkum :)

Umut