SF2568 Parallel Computations for Large- Scale Problems

• English B / English 6
• Completed basic course in numerical analysis (SF1544, SF1545 or equivalent) and
• Completed basic course in computer science (DD1320 or equivalent).

• Basic ideas including hardware architectures, memory hierarchies, communications,
  parallelization strategies, measures of efficiency;
• HPC and Green Computing;
• Introduction to MPI, the Message Passing Interface;
• Simple numerical algorithms including matrix operations, Gaussian elimination;
• Algorithms on graphs including graph partitioning problems;
• Parallel sorting;
• More advanced parallel algorithms;
• Standard libraries;

The goal of the course is to provide a basic understanding of how to develop algorithms and how to implement them in distributed memory computers using the message-passing paradigm.

After completion of the course components the student shall be able to:

• select and/or develop algorithms and data structures for solving a given problem after having analyzed and identified properties of the problem which have the potential for an efficient parallelization;
• theoretically analyze a given parallel algorithm with respect to efficiency and afterwards experimentally evaluate a program for parallel computing by running it on a high-performance computer;
• implement a given algorithm on a distributed-memory computer using the message passing library MPI;
• independently solve a more complex problem and present the results both orally and in writing in a scientific manner;
• identify challenges of Green Computing in HPC.