Showing results for

- Intel Community
- Software Development SDKs and Libraries
- Intel® oneAPI Math Kernel Library & Intel® Math Kernel Library
- Finding Intel Math Kernel Library Implementations of IMSL functions

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Mute
- Printer Friendly Page

Highlighted
##

Baba_Yara__Fahiz

Beginner

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

11-01-2017
10:30 AM

37 Views

Finding Intel Math Kernel Library Implementations of IMSL functions

Tangentially, is a structured way of going about finding high quality free implementations of these subroutines online?

2 Replies

Highlighted
##

mecej4

Black Belt

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

11-01-2017
06:19 PM

37 Views

It is unlikely that you will find an MKL equivalent subroutine with the same argument list and functionality as an IMSL subroutine, in general. That is probably why there is no widely known crossover list.

For normally distributed random numbers, use **vRngGaussian**, see https://software.intel.com/en-us/mkl-developer-reference-c-vrnggaussian#40716A0C-A831-465F-9285-D28DFA06C46A .

Highlighted
##

Zhen_Z_Intel

Employee

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

11-01-2017
08:00 PM

37 Views

Hi,

Seems DRNNOA uses nkira method, actually MKL PRNG based on Gaussian distribution only provide box-muller (which is not acceptance-rejection tech) and inverse DCF (which probably same as DNNOR).

In fact, the ISML numerical lib can be optimized with Intel MKL, some basic function may replaced to implemented with MKL routines. We probably could check if the particular function DRNNOA can be improved by MKL. Thanks.

Best regards,

Fiona

For more complete information about compiler optimizations, see our Optimization Notice.