Speed Up Math Computations on GPUs with Intel® oneMKL

Speed Up Math Computations on GPUs with Intel® oneMKL

@IntelDevTools

Subscribe Now

Stay in the know on all things CODE. Updates are delivered to your inbox.

Sign Up

Overview

Learn how to speed up computation on GPUs using Intel® oneAPI Math Kernel Library (oneMKL), an award-winning, high-performance library that contains optimized threaded and vectorized math routines used for science, engineering, and financial applications.

Topics covered include:

  • An overview of oneMKL and its current support coverage across different domains.
  • Elemental methods to offload oneMKL tasks on GPUs, including using OpenMP* offload, SYCL* buffers, and SYCL interfaces.
  • Implicit and explicit scaling on multistack GPUs.

Discussion points are elaborated with sample code.

Skill level: Intermediate

 

Featured Software

Download the stand-alone version of oneMKL or as part of the Intel® oneAPI Base Toolkit.
 

Code Samples

Download a variety of samples on GitHub* for oneMKL, including:

  • Matrix Multiply: Shows how to use the oneMKL optimized matrix multiplication routines.
  • Fourier Correlation: Learn how to implement 1D and 2D Fourier correlation using SYCL, oneMKL, and Intel® oneAPI DPC++ Library (oneDPL) kernel functions.
  • Monte Carlo European: Shows how to use the library’s random-number generation (RNG) functionality to compute European option prices.

View All

Jump to:

You May Also Like
 

Intel® oneAPI Math Kernel Library

Accelerate math processing routines and increase performance with advanced math routines and functions for science, engineering, or financial applications.

 

 

See All Tools