School of
Engineering sciences

SF2568, Parallel Computations for Large-Scale Problems, 7.5 ECTS

The course will be given at KTH during the academic year 2018/2019. The first lecture is

Tuesday, January 15, 10-12
Lecture Room D41

More details can be found on the course's home page.

General Description

During the last decades we observed a very fast increase in the performance of processors thus enabling to solve larger and larger problems in a reasonable amount of time. However, the speed of computation is limited by the speed of light and the molecule sizes. The way out is parallel computation: Hardware manufactures integrate more and more processing units on a single die.

Parallelization helps to overcome also other bottlenecks of computations, for example:

  • The gap between the speed of the memory/disk subsystem and that of the processors becomes increasingly larger;
  • Data appear distributed (internet resources, distributed data bases etc.)

The present course is intended to provide an introduction into the basic ideas and methods used for developing parallel applications. We will concentrate on distributed memory architectures. Applications include simple numerical algorithms, image processing, sorting, algorithms on graphs, as well as more advanced numerical techniques.

A great deal of work will be spent in hands-on exercises. For those who are not comfortable with C or Fortran, a short introduction will be provided.

If you are missing the first lecture, you will not get access to the computer ressources at PDC.

Further information about the course is given by

Michael Hanke,, (course leader).

Course evaluation

By the end of the course I would like to ask you to answer the questions of the course evaluation form. You will then find a question-form here.

Please fill the course evaluation form. Thank you.

You'll find it here:

Current Course-information

Published by: Michael Hanke <>
Updated 2019-05-14