Syllabus for High Performance and Parallel Computing

Learning outcomes

On completion of the course the student shall be able to:

• implement computational algorithms to efficient code for modern computer architectures;
• use tools for performance optimisation and debugging;
• propose and implement efficient performance optimisations;
• identify factors that restrict parallelism in an algorithm or a program;
• analyse opportunities for sharing and duplication of data within the memory hierarchy;
• present written performance analysis in a clear and explicit way.

Content

Introduction to high performance programming and hardware. Different types of computer architectures and memory organisations. Important concepts such as data distribution, load balancing, locking and syncronization.

Parallel programming with native operating system threading as well as OpenMP, including support for heterogeneous architectures. Analysing bandwith and latency when designing data flows. Shared memory synchronization concepts, including mutexes and atomic operations. Task based programming. Tools and methods for problem solving, software development, debugging, performance analysis and performance optimization.

Instruction

Lectures, computer labs, assignments and projects.

Assessment

Assignments and projects reported both as written reports and orally.

If there are special reasons for doing so, an examiner may make an exception from the method of assessment indicated and allow a student to be assessed by another method. An example of special reasons might be a certificate regarding special pedagogical support from the disability coordinator of the university.

Other directives

The course cannot be included in the same degree as 1DL560 and 1TD351.