CMSC 411 Project
A General Overview of Parallel Processing
Fall 1998
A General Overview of Parallel Processing
Mark Pacifico mjpac@wam.umd.edu Mike Merrill mwm@glue.umd.edu
KEYWORDS
  • Parallelism
  • Interprocessor Communication
  • Synchronization
  • SIMD/MIMD
ABSTRACT
Parallel Processing is the use of multiple processors to co-operate on the solution to a single problem.  Computing in this way vastly speeds up execution time and is necessary in solving massively complex problems such as climate prediction or galactic evolution.  The focus of this project will be on how parallel processing works and how it is used in today's computing world.  Topics that will be touched upon include the history of parallel processing, the advantages and drawbacks to using parallel architectures, and how parallel processing might evolve over the next few years.
WHAT THE READER SHOULD TAKE AWAY
The reader of the web page should learn more about parallel processing and hopefully gain a better understanding of chapter 8 (Multiprocessors) in Hennessy and Patterson.  We would also like to include some problems dealing with how parallel processing can speed up execution time.  The reader should get some practice doing these problems and that will help prepare the reader for the final exam.
OUTLINE
  • Introduction
    • Parallel processing - definition
    • Purpose
    • Thesis statement
  • History
    • Past needs for parallel computing
    • Changes in parallel computing
    • Present/future applications for parallel computing
  • Parallelism
    • SIMD
    • MIMD
  • Communication
    • Shared/distributed memory
    • Communication cost
    • Latency/bandwidth
  • Synchronization
    • Mutual exclusion
    • Barriers
    • Memory consistency
  • Summary/Future trends of parallel processing
  • Practice questions/answers
REFERENCES
Proposal | Introduction | History | Parallelism | Communication | Synchronization | Summary & Questions
Copyright 1998, Mark Pacifico & Mike Merrill