The demand for ever-increasing computational power drives the development and exploitation of high-performance computing that underpins leading edge research in computationally intensive scientific and engineering fields. This module is designed to equip you with a solid foundation and useful skills in high-performance computing. In this module you will learn about current high-performance computer architectures and how the computer architecture influences the performance of algorithms and programs. You will also develop skills in parallel algorithm design and parallel programming, and will gain experience of using a high-performance computing system.

PRE-REQUISITE MODULES: ECM1416, ECM2433
 

This module aims to provide you with a thorough grounding in parallel programming and the architectures used in high-performance computing. After presenting the fundamental ideas and basic concepts of high-performance computing, the module outlines the architectures, components and parallel programming of high-performance computers. The module will introduce you to recent developments and future trends in architecture and algorithms in high-performance computing.

On successful completion of this module you should be able to:

Module Specific Skills and Knowledge

1. demonstrate an understanding of the fundamental ideas and issues of high-performance computing;

2. demonstrate knowledge of high-performance computer architectures;

3. demonstrate skills in parallel processing algorithm design and the practical implementation of such algorithms;

4. demonstrate knowledge of how to quantitatively assess the performance of parallel programs;

5. demonstrate an awareness of numerical effects and the influence of floating point number representation in high-performance computing applications.

Discipline Specific Skills and Knowledge

6. Understand how computer architectures can influence the performance of algorithms and programs.

7. interpret an informal requirement specification;

8. systematically analyse information and make appropriate design choices.

Personal and Key Transferable / Employment Skills and Knowledge

9. relate reading materials to lecture content and analyse a problem;

10. use technical manuals and books to interpret technical errors.

 

 

- motivation of and introduction to high-performance computing;

- parallel computer architecture: shared-memory and distributed-memory architectures, multi-core processors, Graphics Processing Units (GPUs);

- interconnection networks in high performance computers: topologies, latency and bandwidth;

- parallel processing algorithm and programming design:, domain decomposition, halo exchange, manager-worker, task-based parallelism;

- parallel programming methods: Message Passing Interface (MPI), OpenMP

- parallel performance: speed-up, efficiency, parallel overheads and scaling;

- floating point arithmetic: floating point model, range, accuracy, exceptions.
 

Reassessment will be by coursework and/or written exam in the failed or deferred element only. For referred candidates, the module mark will be capped at 40%. For deferred candidates, the module mark will be uncapped.

Basic reading:

ELE:  http://vle.exeter.ac.uk/

Web based and Electronic Resources:

Other Resources:

Reading list for this module: