Showing results for

- Intel Community
- Software Development SDKs and Libraries
- Intel® oneAPI Math Kernel Library & Intel® Math Kernel Library
- no consistent result with mkl 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
##

Is anyone can solve the problem that mkl functions didn't get consistant result?

Emily_Zhu

Beginner

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

05-13-2011
10:14 AM

17 Views

no consistent result with mkl functions

we use 10.3.3, It looks any function like DGEMM, DSYEV get different results

The different is big since for 1000*1000 matrix, the max difference for single element is

the matrix is 1.0e-14.

We also try 16 boundary alignment, but it doesn't work.

4 Replies

Highlighted
##

We've generally seen that if you are running your program on the same computer, with the same number of threads and your memory is aligned on 16-byte boundaries then you'll get consistent results from DGEMM from one run to the next.

We'd need to know a little more about your program and the system on which you are running.

Todd_R_Intel

Employee

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

05-13-2011
04:46 PM

17 Views

We'd need to know a little more about your program and the system on which you are running.

Highlighted
##

If I set each computer as mkl_sequential and use 16 boundary alignment.

Emily_Zhu

Beginner

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

05-15-2011
11:47 PM

17 Views

but send whole computation to cluster, if the final result will be same?

Highlighted
##

*If I set each computer as mkl_sequential and use 16 boundary alignment.*If each node has the same software installation and same CPU brand, it's reasonable to hope for identical results.

TimP

Black Belt

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

05-16-2011
07:16 AM

17 Views

Quoting Emily Zhu

but send whole computation to cluster, if the final result will be same?

Highlighted
##

Emily_Zhu

Beginner

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

05-16-2011
11:42 AM

17 Views

Another question is that alought DGEMM is repeatable, if the result will shift with real result of matrix

multiplication. And how much it will shift?

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