Intel® oneAPI Math Kernel Library
Ask questions and share information with other developers who use Intel® Math Kernel Library.
6977 Discussions

Join the Intel® Parallel Studio XE 2019 Beta Program today

Shaojuan_Z_Intel
Employee
431 Views

Join the Intel® Parallel Studio XE 2019 Beta Program today and—for a limited time—get early access to new features and get an open invitation to tell us what you really think.

We want YOU to tell us what to improve so we can create high-quality software tools that meet your development needs.

Sign Up Now >

Top New Features in Intel® Parallel Studio XE 2019 Beta

  • Scale and perform on the path to exascale. Enable greater scalability and improve latency with the latest Intel® MPI Library.
  • Get better answers with less overhead. Focus more fully on useful data, CPU utilization of physical cores, and more using new data-selection support from Intel® VTune Amplifier’s Application Performance Snapshot.
  • Visualize parallelism. Interactively build, validate, and visualize algorithms using Intel® Advisor’s Flow Graph Analyzer.
  • Stay up-to-date with the latest standards:
    • Expanded C++17 and Fortran 2018 support
    • Full OpenMP* 4.5 and expanded (partial) support for OpenMP* 5.0 (Preview 2) specification
    • Python* 3.6 and 2.7

New Features in Intel® Math Kernel Library

  • Introduced automatic S/DGEMM JIT capability for small matrix sizes (m,n,k <=16) to improve S/DGEMM performance for Intel® Advanced Vector Extensions 2 (Intel® AVX2)  and Intel® Advanced Vector Extensions 512 (Intel® AVX-512)  when compiling with MKL_DIRECT_CALL_JIT (threaded usage) or MKL_DIRECT_CALL_SEQ_JIT (sequential usage).
  • Introduced new functions to JIT (create) optimized S/DGEMM-like matrix multiply kernels for small matrix sizes (m,n,k <=16) for Intel® Advanced Vector Extensions 2 (Intel® AVX2)  and Intel® Advanced Vector Extensions 512 (Intel® AVX-512), execute the optimized kernel created using matrices with matching dimensions, and to remove (destroy) the JIT kernel.
  • Introduce SparseQR functionality
  • Introduced Extreme{EVD/SVD} functionality
  • Introduced Multinominal Random Number Generators

To learn more, visit Intel® Parallel Studio XE 2019 Beta page.

Then sign up to get started.

Note: If you encounter a HTTP ERROR: 404 on the beta registration page, please clear the browser cache and try again.

0 Kudos
5 Replies
JohnNichols
Valued Contributor III
431 Views

I updated to the new BETA.  But when I compile my code I get a new error -- 

warning #13046: system locale 3081 is not supported and reverting to English messages.

Any ideas on how to fix it 

0 Kudos
mecej4
Honored Contributor III
431 Views

Locale 3081 corresponds to Australian English. It is only a warning, and the differences probably do not matter.

0 Kudos
JohnNichols
Valued Contributor III
431 Views

Dear mecej4:

Thanks for the note. I do use Australian English. Interesting the things they fix in Fortran with new releases. 

John

0 Kudos
Chuck1
Beginner
431 Views

Hi,

The program doesn't start correctly- it gives following error when run "mpivars" . Both command prompts

All programs-> Command Prompt with Intel Compiler 19.0 Beta-> Intel 64 Visual Studio 2015 environment

All programs-> Command Prompt with Intel Compiler 19.0 Beta->Intel 64 Build Environment

Do you know this error?

fi_info.exe-Application Error

intel_error_.png

My system:

Intel® Parallel Studio XE 2019 Beta

Intel® Xeon® Processor E5-2687W, Win 7 Enterprise

0 Kudos
Royi
Novice
431 Views

Is there a blog post about those JIT improvements?

0 Kudos
Reply