topic Re:The unexpected result when using Pardiso solver to solve linear equations in parallel (OpenMP) in Intel® oneAPI Math Kernel Library

The unexpected result when using Pardiso solver to solve linear equations in parallel (OpenMP)

Ewigstern — Thu, 08 Sep 2022 12:41:50 GMT

Dear Intel Team,

I recently found a strange phenomenon when invoking the pardiso solver to solve a linear equation, the details are as follows:

For a linear equation with multiple right-hand sides: AX=B, where A is an nxn matrix, B is an nxm matrix, and m is the number of right-hand sides.

When I solve the equation in the following way (pseudocode), the result is correct.

do i=1,m
call pardiso(A,B(i),X(i))
end do

correct result:

However, when I solve the equation in the following way (pseudocode), all or some of the threads compute the same result, i.e. all or some of the columns of X are the same. Notice:
1. I have confirmed that the B(i) entered in each loop is correct.
2. These same calculation results may be the correct result of a certain thread, or may be the initial 0 value.
3. In rare cases, all threads can get different calculation results, and the results are correct.

!$OMP Parallel Do Num_Threads(p)
do i=1,m
call pardiso(A,B(i),X(i))
end do
!$OMP end Parallel Do

wrong result:

Such a phenomenon is very strange, and it confuses me. I would like to know what is causing this? Is it not possible to call the pardiso solver in the OpenMP way? Or do I need to adjust the input parameters of the pardiso solver when calling the pardiso solver in OpenMP?

Best wishes,
Ewigstern

Re:The unexpected result when using Pardiso solver to solve linear equations in parallel (OpenMP)

VidyalathaB_Intel — Fri, 09 Sep 2022 11:21:49 GMT

Hi Ewigstern,

Thanks for reaching out to us.

Could you please provide us with the following details so that we could try reproducing the issue from our end?

> MKL Version being used

> Complete sample reproducer code and steps to reproduce it (commands to compile and run)

> OS environment details

> Expected results and observed results (i.e. the results that you are getting while running in parallel mode)

Regards,

Vidya.

Re:The unexpected result when using Pardiso solver to solve linear equations in parallel (OpenMP)

VidyalathaB_Intel — Wed, 14 Sep 2022 13:07:22 GMT

Hi Ewigstern,

Since this is a duplicate thread of https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/The-wrong-result-when-using-Pardiso-solver-to-solve-linear/td-p/1414332, we will no longer monitor this thread. We will continue addressing this issue in the other thread.

Regards,

Vidya.