Reply to topic

Eigen Dense Performance on iOS/Android + BLAS / LAPACK

rshauty
Registered Member
Posts
3
Karma
0
https://eigen.tuxfamily.org/dox/TopicUs ... apack.html

"Since Eigen version 3.3 and later, any F77 compatible BLAS or LAPACK libraries can be used as backends for dense matrix products and dense matrix decompositions. For instance, one can use Intel® MKL, Apple's Accelerate framework on OSX, OpenBLAS, Netlib LAPACK, etc."

What is the general expectation of using one of these libraries instead of Eigen itself as the backend for dense operations when running on Android and when running on iOS? I understand that it is possible, and that "go test it for your scenario" is definitely in effect here. However, as many other literature on Eigen from its own site and across the web reports, Eigen should/can be faster than using these libraries as the backend for dense operations.

Wondering if anyone has tested and has seen evidence one way or another, specifically for ARM platforms, but also curious about OSX.

Thanks!

~R

Reply to topic

Bookmarks



Who is online

Registered users: Baidu [Spider], Bing [Bot], claydoh, Exabot [Bot], Google [Bot], jackdinn, La Ninje, rolfreiner, slowersu, Sogou [Bot], vinnywright, Yahoo [Bot], Zeznon