Technology trends have lead to the introduction of many-core processors in all domains of computing, leading to the inevitable consequence that parallel programs must be developed. In this talk, we first walk through an advanced example of parallel programming, namely programming the Cell B.E. processor and we point out why this is not so easy. Then, we present a first step towards making parallel programming easy for a large group of programmers. This approach uses implicit parallel programming and targets homogeneous shared-memory multi-cores
Hans Vandierendonck is a Post-Doctoral Research Fellow with the Fund for Scientific Research - Flanders. He is associated to Ghent University. His research interests include parallel programming of multi-core processors, programming models and auto-parallelization. Prior to this, he has worked on computer architecture, in particular cache memories and branch prediction. He obtained the degree of Engineering at Ghent University and the PhD degree at the same university.