Module 02512 (2001)
Syllabus page 2001/2002
06-02512
Parallel Computer Architecture
Level 3/H
Links | Outline | Aims | Outcomes | Prerequisites | Teaching | Assessment | Books | Detailed Syllabus
The Module Description is a strict subset of this Syllabus Page. (The University module description has not yet been checked against the School's.)
Relevant Links
Outline
The wide variety of parallel computer architectures in use today is examined. The architectural categories studied include SIMD, shared-memory MIMD and message-passing MIMD. Some basic parallel programming techniques are also discussed, along with the effect of the architecture on programming and the use of abstract models such as PRAM.
Aims
The aims of this module are to:
- to introduce the main classes of parallel computer architectures and their distinguishing characteristics
- to examine some examples of particular architectures that were important in the development of parallel computers
- to examine some examples of recent parallel architectures
- to introduce some main principles of parallel programming and the effects of the architecture on algorithm design
Learning Outcomes
| On successful completion of this module, the student should be able to: | Assessed by: | |
| 1 | understand the main classes of parallel architectures and their distinguishing characteristics and relative merits | Examination |
| 2 | understand the major characteristics of some examples of the main architectural classes | Examination |
| 3 | design and analyse simple parallel algorithms | Examination |
| 4 | understand the impact of parallel computer architecture on software design and performance | Examination |
Restrictions, Prerequisites and Corequisites
Restrictions:
None
Prerequisites:
06-08752 (Introduction to Software & Hardware Engineering A)
Co-requisites:
None
Teaching
Teaching Methods:
2 hrs lectures/tutorials per week
Contact Hours:
Assessment
- Supplementary (where allowed): As the sessional assessment
- 2 hr examination (100%).
Recommended Books
| Title | Author(s) | Publisher, Date |
| Lecture Notes | Axford, T H | |
| Highly Parallel Computing (2nd ed.) | Almasi & Gottlieb | Benjamin Cummings, 1994 |
Detailed Syllabus
-
SIMD Array Processors (4 lectures)
- Example architectures: Illiac IV, AMT DAP, Connection Machine, systolic arrays.
- Interconnection Networks (3 lectures)
- Basic characteristics, routing functions, network topologies, dynamic (switched) networks (cross-bar, bus, Benes networks, shuffle-exchange networks). Example: Connection Machine.
- MIMD Shared-Memory Architectures (3 lectures)
- Example architectures: BBN Butterfly, DEC AlphaServer 8000 as a symmetric multiprocessor architecture. Cache coherence protocols.
- MIMD Message-Passing Architectures (2 lectures)
- The Transputer as a building block, complete machines using Transputers, hypercube machines (Cosmic Cube, Intel Hypercube and hypercubes using Transputers).
- Programming Parallel Machines (5 lectures)
- The PRAM Model, the BSP Model, writing parallel programs for abstract machines, the divide-and-conquer paradigm and some example algorithms, the doubling paradigm and example algorithms, programming real machines (SIMD, SPMD, shared-memory MIMD, and message-passing MIMD), conclusions and future directions.
- Discussion of Exercises (remaining lectures)
Last updated: 8 January 2002
Source file: /internal/modules/COMSCI/2001/xml/02512.xml
Links | Outline | Aims | Outcomes | Prerequisites | Teaching | Assessment | Books | Detailed Syllabus