Postgraduate Course: Performance Programming (EPCC11009)
Course Outline
| School | School of Informatics |
College | College of Science and Engineering |
| Credit level (Normal year taken) | SCQF Level 11 (Postgraduate) |
Availability | Available to all students |
| SCQF Credits | 10 |
ECTS Credits | 5 |
| Summary | Performance is key for both efficient use of computing hardware and ensuring applications (be they computational simulation, machine learning, databases, etc...) complete their tasks in an acceptable amount of time. However this is one of the more difficult requirements to satisfy:
- Issues effecting performance often vary between different hardware and software environments. This requires performance issues to be frequently re-visited as the hardware and software environment changes.
- Performance programming requires detailed knowledge of the underlying environment
- The design decisions necessary to achieve good performance are often in conflict with other desirable properties of the program.
After taking this course students should have a good practical understanding of the general issues and methodologies associated with designing building and refactoring codes to meet performance requirements. In addition they will have an overview of a number of subjects that are important in the understanding of performance on current systems. |
| Course description |
The course will cover the following topics:
- Overview of performance programming. Methodology, the optimisation cycle.
- Designing for performance. Encapsulation as an aid to performance tuning.
- Tools for performance programming. Profilers and code instrumentation.
- Compilers and compiler optimisation.
- Memory hierarchies, Memory structures and associated optimisations.
- Performance tuning for shared memory.
- Floating point performance. Pipelines, SIMD, vectorisation.
|
Entry Requirements (not applicable to Visiting Students)
| Pre-requisites |
It is RECOMMENDED that students have passed
HPC Architectures (EPCC11004) OR
HPC Architectures (INFR11175)
|
Co-requisites | |
| Prohibited Combinations | |
Other requirements | Students are required to be familiar with C/C++ or Fortran for practical exercises and in-class examples. It is strongly RECOMMENDED that students have passed HPC Architectures or an equivalent course. |
Information for Visiting Students
| Pre-requisites | Students are required to be familiar with C/C++ or Fortran for practical exercises and in-class examples. It is strongly RECOMMENDED that students have passed HPC Architectures or an equivalent course. Familiarity with HPC systems from other EPCC courses will be beneficial. |
Course Delivery Information
|
| Academic year 2023/24, Available to all students (SV1)
|
Quota: 84 |
| Course Start |
Semester 2 |
Timetable |
Timetable |
| Learning and Teaching activities (Further Info) |
Total Hours:
100
(
Lecture Hours 21,
Supervised Practical/Workshop/Studio Hours 11,
Feedback/Feedforward Hours 1,
Programme Level Learning and Teaching Hours 2,
Directed Learning and Independent Learning Hours
65 )
|
| Assessment (Further Info) |
Written Exam
0 %,
Coursework
100 %,
Practical Exam
0 %
|
| Additional Information (Assessment) |
A single coursework consisting of two submissions (25%/75%).
Coursework is expected to be worked on consistently across the Semester building on material from lectures and practicals as they are delivered. Although the first and second submission dates are sequential they are inherently linked and some work not assessed until the second submission is clearly relevant to feed into the first submission. The two should be treated as a single coherent piece of work with the first coursework offering students an opportunity to receive feedback which can be acted upon to improve the later higher-stakes submission. |
| Feedback |
Provided via the practical classes and on assessed work in preparation for next hand-in. |
| No Exam Information |
Learning Outcomes
On completion of this course, the student will be able to:
- Understand and demonstrate the appropriate methodology when attempting to improve code performance.
- Understand how performance is achieved via hardware, compilers and operating systems.
- Appreciate the limitations of systems and recognise when these will have a serious impact.
- Interpret the observed performance of code in terms of how its execution is realised on the system.
- Identify code regions appropriate for manual optimisation and propose, implement and evaluate optimisations on these regions.
|
Reading List
| Provided via Learn/Leganto |
Additional Information
| Graduate Attributes and Skills |
Programming
Optimisation
Solution Exploration and Evaluation
Communication of technical information via written report
Time management |
| Special Arrangements |
There are limited spaces on this course. Students not on the MSc in High Performance Computing or MSc High Performance Computing with Data Science should contact the course secretary to confirm availability and confirm that they have the required prerequisites before being enrolled on the course. |
| Keywords | Performance Programming,HPC,EPCC,Parallelism,Optimisation,Profilers,Compilers,SIMD,Vectors |
Contacts
| Course organiser | Mr William Jackson
Tel:
Email: Adrian.Jackson@ed.ac.uk |
Course secretary | Mr James Richards
Tel: 90131 6)51 3578
Email: J.Richards@epcc.ed.ac.uk |
|
|
University Homepage