Class Descriptions

Please note that not every class is offered every quarter, but most classes can be made available to departments on request. You can contact atshpc@ucla.edu to arrange a special presentation.

High Performance Computing

Parallel Computing Using MPI-Message Passing Interface

This class has been designed to familiarize you with Parallel Computing using MPI. At the end of this three-session class you will be able to write simple parallel programs. This will allow you to take advantage of the parallel nature of a Beowulf cluster, such as the one available to you at ATS, or one that your department or professor might have. In today's computing environment, Beowulf clusters, built with off-the-shelf commodity components, are the only economical way to run large codes in a reasonable time.

In order to understand the topics presented in this class, you should have some experience in programming with Fortran or C and should plan to attend all the sessions, starting with the first.

First Session - Introduction to Parallel Computing (Schedule)

This session is designed to introduce you to the concepts of parallel computing. No prior knowledge in High Performance Computing is required. It introduces basic topics, such as the nature of parallelism; underlying terminology, including bandwidth and latency, pipelining and cache; and different types of parallel systems, notably shared memory and distributed memory clusters. All the concepts that you need to know in order to program with MPI will be discussed. These include communication strategies using point to point and broadcast communications, data decomposition, and performance evaluation. The class will focus on use of beowulf clusters, a types of parallel system commonly deployed at UCLA. (2 hours)

Second Session - Parallel Programming using MPI (Schedule)

This two-hour session will cover the use of MPI in the implementation of parallel codes. It will be a class-room session which will introduce the basic functionality of MPI and show how to use it in writing parallel codes designed to run on clusters. It will introduce the syntex of the basic MPI calls and it will cover their use of them with the help of numerous examples. (2 hours)

Third Session - Hands on with MPI (Schedule)

The third session is a hands-on session where you will gain practical experience with MPI. You will first be given example codes to compile, run, and modify. All codes will be run on the beowulf cluster at ATS. Instructors will be on hand to assist you. You will be allowed to write a code of your choice in order to better understand the topic. (2 hours)

After the class is over, you may continue working on your parallel programs and ATS will assist you write a parallel program, convert a serial program into a parallel program, or to speed up a program which isn't performing appropriately.

At any time, you can make an appointment to see one of the ATS consultants by writing to atshpc@ucla.edu