High performance computing with MATLAB

Version 1.0.0 (15.3 MB) by Huming
Slides 1、introduction 2、cluster computing I 3、parallel computing 4、MPI 5、GPU 6、HPC+AI 7、Project
216 Downloads
Updated 6 Jun 2022

View License

Course
Description:
This is a hands-on training course to learn the basic skills to accelerate your MATLAB programs using the MATLAB Parallel Computing Toolbox on regular desktop computers, GPU enabled computers, High-Performance Computing clusters, and cloud-based MATLAB parallel computing environment.
近年来,由于频率墙和功耗墙的存在,计算机计算性能的提升主要依赖于计算核心数量的增加,这使得传统的串行算法设计逐步转向基于多核和众核的并行算法设计,越多越多的应用领域,特别是人工智能领域需要并行数据结构与算法设计的相关基础。本课程介绍了现代计算机CPUGPU硬件架构的发展,介绍了多核与众核的并行编程技术,给出了并行程序的设计方法,详细阐述了MATLAB并行计算工具箱的使用方法并给出了程序示例。最后给出了基于MATLAB的并行算法设计例子。
Ⅰ理论部分
第1章 概述 (2学时)
高性能计算的意义,国内外研究进展;能够解决的科学和工程问题。
第2章 高性能计算平台 (2学时) (迈斯沃克公司可提供最新技术和平台资料)
迈斯沃克并行计算体系结构介绍。
第3章 并行程序设计模型与性能评价(2学时)
并行程序设计方法PCAM,Amdahl定律,Gustafson定律,加速比等。
第4章 基于消息传递编程(MPI)的并行程序开发 (2学时)
MPI并行程序设计开发,点对点通信,阻塞和非阻塞通信,聚合通信。
第5章 循环并行化(2学时)
Parfor,并行化开销,嵌套训练的并行化。
第6章 MATLAB与GPU计算 (2学时)
GPU软硬件架构,MATLAB使用GPU。
第7章 高性能计算与大数据处理和深度学习 (2学时)
为大数据创建数据存储(datastores),tall数组;MATLAB平台上深度学习算法与高性能计算。
Instructor Biography
Huming Zhu received his Bachelor degree in Electronic Engineering, Master degree and Ph.D. degree in Communication and Information system from Xidian University, Xian, China, in 2001, 2004 and 2010,respectively. He is an associate professor at Xidian University. His research interests mainly include data mining,
pattern recognition, and High Performance Computing and image processing.

Cite As

Huming (2024). High performance computing with MATLAB (https://www.mathworks.com/matlabcentral/fileexchange/112740-high-performance-computing-with-matlab), MATLAB Central File Exchange. Retrieved .

MATLAB Release Compatibility
Created with R2020a
Compatible with any release
Platform Compatibility
Windows macOS Linux
Tags Add Tags

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!
Version Published Release Notes
1.0.0