Postgraduate Course: Parallel Decomposition (PGPH11006)
Course Outline
School |
School of Physics and Astronomy |
College |
College of Science and Engineering |
Course type |
Standard |
Availability |
Not available to visiting students |
Credit level (Normal year taken) |
SCQF Level 11 (Postgraduate) |
Credits |
10 |
Home subject area |
Postgraduate (School of Physics and Astronomy) |
Other subject area |
None |
Course website |
http://www.epcc.ed.ac.uk/msc/ |
|
|
Course description |
The first section of this course covers the use of the data parallel programming paradigm. In its simplest form the data-parallel programming model extends a sequential programming language with parallel constructs for handling large aggregates of data such as arrays. The course focuses on the extension of Fortran 90 to High Performance Fortran (HPF), the de facto standard for data parallel languages and includes extensive practical exercises.
The final section of this course describes the main classes of techniques used to decompose an application across a parallel computer. The course covers each technique in some detail and discusses when it can be used effectively. |
Entry Requirements
Pre-requisites |
|
Co-requisites |
Students MUST also take:
Fundamental Concepts of High Performance Computing (PGPH11014)
|
Prohibited Combinations |
|
Other requirements |
Only postgraduates or final year MPhys students with suitable computational background, subject to space restrictions and agreement with relevant Programme Coordinator.
|
Additional Costs |
None |
Course Delivery Information
|
Delivery period: 2010/11 Semester 1, Not available to visiting students (SS1)
|
WebCT enabled: Yes |
Quota: None |
Location |
Activity |
Description |
Weeks |
Monday |
Tuesday |
Wednesday |
Thursday |
Friday |
No Classes have been defined for this Course |
First Class |
First class information not currently available |
Summary of Intended Learning Outcomes
On completion of this module students should be able to:
- Demonstrate understanding of data parallel programming.
- Implement data parallel codes using F90 array syntax and HPF directives.
- Describe the main approaches to parallel decomposition.
- Evaluate and choose appropriate decompositions. |
Assessment Information
100% examination consisting of a two hour exam |
Special Arrangements
Not entered |
Contacts
Course organiser |
Dr Judy Hardy
Tel: (0131 6)50 6716
Email: j.hardy@ed.ac.uk |
Course secretary |
Yuhua Lei
Tel: (0131 6) 517067
Email: yuhua.lei@ed.ac.uk |
|
copyright 2010 The University of Edinburgh -
1 September 2010 6:29 am
|