Intel® oneAPI Math Kernel Library
Ask questions and share information with other developers who use Intel® Math Kernel Library.
Announcements
Welcome to the Intel Community. If you get an answer you like, please mark it as an Accepted Solution to help others. Thank you!

FEA and CFD with Math Kernel Library

Dinamitry
Beginner
216 Views

Good day!

I present R&D department of organisation. We use Intel Core i7-6700, 7700 CPUs. 

Our main software for ingineering calculations and simulations are Autodesk Inventor 2021.1, Autodesk CFD 2021.1. 

1 st question is: can Intel® oneAPI Math Kernel Library(link below) installation speed up FEA calculations, CFD calculations?

2 nd question is a bit more wider. We know from the practice of using various engineering software that they all are diffrent in multicore usage possibility. For example Solidworks Flow Simulation uses all avalable cores, but Autodesk CFD just sometimes. Finally, in task manager there is 17-60% usage of CPU for Autodesk CFD and 100% in any moment of simulation for SolidWorks Flow Simulation. 

So, Math Kernel Library can operates well and can be helpfull only for software which supports multicore usage, or its can be helpfull in any case?

 

Math Kernel Library link

https://software.intel.com/content/www/us/en/develop/tools/oneapi/components/onemkl.html

 

Thank You in advance!

0 Kudos
1 Solution
mecej4
Black Belt
202 Views

MKL consists of a collection of library routines, implemented in static or shared libraries, and files to facilitate compiling and linking user or third party code with these libraries.

If third-party software detects the presence of MKL and has built-in capabilities to issue calls to the MKL routines, that software may run faster and use multicore capabilities more efficiently. Otherwise, simply installing MKL will have no effect on the third party software.

View solution in original post

2 Replies
mecej4
Black Belt
203 Views

MKL consists of a collection of library routines, implemented in static or shared libraries, and files to facilitate compiling and linking user or third party code with these libraries.

If third-party software detects the presence of MKL and has built-in capabilities to issue calls to the MKL routines, that software may run faster and use multicore capabilities more efficiently. Otherwise, simply installing MKL will have no effect on the third party software.

View solution in original post

Dinamitry
Beginner
198 Views

Thank You for information!

Reply