Supercomputers for starters

 This is the second course in a series of three. In the first two-session course (February 10 and 13, 2020), we look at Linux, the operating system used on all VSC supercomputers. In this one-lecture course, we take a closer look at the architecture of a supercomputer and the consequences that this has for running and developing software. At the hardware level, we explore the (many) similarities and (fewer but crucial) differences with PCs and smartphones. We also take a brief look at the so-called middleware and programming programming models for supercomputers without actually teaching supercomputer programming. These sit between the hardware and the application you want to run or develop. This will help you understand the software requirements of applications you might use. Some concepts from the first lecture of the Linux course will be used in this lecture (processes, threads, directories).

This lecture is equally relevant to those who will merely be running existing software as to those who will program on the clusters. Not all software will run efficiently on modern supercomputers and some will not run at all. This lecture tries to give you the necessary insight to evaluate software packages for suitability for supercomputers. It will also teach you some of the terminology that you will come across in the manuals of popular scientific packages.

The third and final two-session course in this series (March 2 and 5, 2020) will then show you how to actually start programs on the cluster at UAntwerpen.