Hits:

Date:2025-07-08

Title:High Performance Intelligent Computing and Experiments

Content:

Intel® Certified Instructor in the DPC++ Essentials course

Teaching:

High Performance Intelligent Computing and Experiments(Spring and Fall Semester, 20**-present),The course is included in the masters programs in the School of Artificial Intelligence. As an interdisciplinary course combining theory and experiments, the course will train students on using oneAPI’s cross-architecture programming approach; how to exploit cutting-edge features of CPUs, GPUs, and other accelerators; and optimizing code using common programming tools that support familiar languages (C++, Fortran, OpenMP, etc.) and standards while providing full continuity with existing code. Building on more than 10 years of collaboration between XDU and Intel in advancing HPC education in China, our goal to develop a leading open source course with experiments based on Intel® oneAPI, to prepare students to become the next generation of HPC programmers and foster HPC educational advancement in China.

Introduction of Intel® oneAPI

With computer system architecture entering a period of prosperity in which CPUs, GPUs, FPGAs, and AI Accelerators complementary to each other, the diversity of hardware brings the complexity of software design and development. Intel® oneAPI is a cross-industry, open, standards-based unified programming model that delivers a common developer experience across accelerator architectures—for faster application performance, more productivity, and greater innovation with its single, unified open development model for performant and productive heterogeneous programming and cross-vendor support.

oneAPI consists of a language and libraries for creating parallel applications:

DPC++: oneAPI’s core language for programming accelerators and multiprocessors. DPCPP allows developers to reuse code across hardware targets (CPUs and accelerators such as GPUs and FPGAs) and tune for a specific architecture

oneDPL: A companion to the DPC++ Compiler for programming oneAPI devices with APIs from C++ standard library, Parallel STL, and extensions.

oneDNN: High performance implementations of primitives for deep learning frameworks

oneCCL: Communication primitives for scaling deep learning frameworks across multiple devices

Level Zero: System interface for oneAPI languages and libraries

oneDAL: Algorithms for accelerated data science

oneTBB: Library for adding thread-based parallelism to complex applications on multiprocessors

oneVPL: Algorithms for accelerated video processing

oneMKL: High performance math routines for science, engineering, and financial applications

Learn more

1.    General introduction

      

2.    Intel oneAPI DevCloud

     http://devcloud.intel.com/oneapi/

3.    Intel Devmesh

     Projects | Intel DevMesh

4.    oneAPI github

     http://github.com/oneapi-src

5.    SYCL Specification

     http://www.khronos.org/sycl/

6.    Intel HPC

     http://www.intel.com/content/www/us/en/analytics/high-performance-computing/overview.html

7.    Intel oneAPI Forum

     http://software.intel.com/en-us/forums/intel-oneapi-forums