OpenMP-Parallelized LU Decomposition Matrix Solver

[JDFI]

Hi,

Is there an OpenMP-Parallelized LU Decomposition Matrix Solver available in MKL library? If yes, is there any guidance link on how to use it?

Thanks,

JDFI

[ShanmukhS_Intel]

Hi,

Thanks for reaching out to us. 

Please refer the below link which could help you regarding OpenMP-Parallelized LU Decomposition Matrix Solver.

https://techdecoded.intel.io/resources/speeding-algebra-computations-with-intel-math-kernel-library-vectorized-compact-matrix-functions/#gs.c3hrjp

Kindly let us know if you need any other information.

Best Regards,

Shanmukh.SS

[JDFI]

Hi,

Thanks for your information. When I clicked your link, it re-directed to another page:

https://software.intel.com/content/www/us/en/develop/tools/oneapi/tech-articles-how-to.html#gs.d4y56b

Thanks,

JDFI

[Gennady_F_Intel]

Yes, LU decomposition routines are threaded by OpenMP. You may take a look at this MKL Dev Guide page: https://software.intel.com/content/www/us/en/develop/documentation/onemkl-linux-developer-guide/top/managing-performance-and-memory/improving-performance-with-threading/openmp-threaded-functions-and-problems.html#openmp-threaded-functions-and-problems_LAPACK

Hot to Link- see a number of examples show how to do that

https://software.intel.com/content/www/us/en/develop/documentation/onemkl-linux-developer-guide/top/linking-your-application-with-the-intel-oneapi-math-kernel-library/linking-examples/linking-on-intel-64-architecture-systems.html

or check MKL Linker Adviser - https://software.intel.com/content/www/us/en/develop/tools/oneapi/components/onemkl/link-line-advisor.html

check "Improving Performance with Threading " chapter from Dev Guide : 

https://software.intel.com/content/www/us/en/develop/documentation/onemkl-linux-developer-guide/top/managing-performance-and-memory/improving-performance-with-threading.html

[JDFI]

Hi,

Thanks for the help. But the example link reports "Error: Page Not Found".

Is there an updated link for this?

Hot to Link- see a number of examples show how to do that

https://software.intel.com/content/www/us/en/develop/documentation/onemkl-linux-developer-guide/top/...

Thanks,

JDFI

[JDFI]

Hi,

The linker adviser page does not generate a workable link. Do I miss anything in my option selections?

Thanks,

JDFI

[ShanmukhS_Intel]

Hi,

Has the solution provided helped. Please let us know if the issue still persists.

Best Regards,

Shanmukh.SS

[JDFI]

Hi ShanmukhS,

The link re-directs to another page. Could you please kindly check it and let me know if there's another workable link?

Thanks,

JDFI

[ShanmukhS_Intel]

Hi,

Thanks for reaching out to us. 

Please refer the below link which could help you regarding OpenMP-Parallelized LU Decomposition Matrix Solver.

https://software.seek.intel.com/LP=18274

Kindly let us know if you need any other information.

Best Regards,

Shanmukh.SS

[JDFI]

Hi,

Could you please send me a download link to get the MKL library which has OpenMP parallelized LU decomposition and Lapack in Fortran?

Thanks,

JDFI

[Gennady_F_Intel]

JDFI

You could find out the link to the download page at the main MKL product page:

https://www.intel.com/content/www/us/en/developer/tools/oneapi/onemkl.html#gs.dh183b

[JDFI]

Hi Gennady,

I downloaded it from the link you provided, but encountered the following error during installation.

Do you have any advice?

Thanks and have a wonderful weekend,

JDFI

[Gennady_F_Intel]

Hi JDFI,

What version of Windows do you use?

[JDFI]

Hi Gennady,

It is Win 7.

Thanks,

JDFI

[Gennady_F_Intel]

ok, I see, thanks. Here is the link to the oneMKL System Requirements - https://www.intel.com/content/www/us/en/developer/articles/system-requirements/oneapi-math-kernel-library-system-requirements.html

Windows*

• Windows* Server 2019
• Windows* 10
• Windows* (HPC) Server 2016 (64-bit version only)

Therefore, the installation doesn't support this OS.

[Gennady_F_Intel]

The thread is closing and we will no longer respond to this thread. If you require additional assistance from Intel, please start a new thread. Any further interaction in this thread will be considered community only.