Higher Performance by less CPUs
Reiner W. Hartenstein · 2006
ISC 2006, Dresden, Germany, June 27 30, 2006 http://www.supercomp.de/ The end of Moore’s Law. Increasing the architectural complexity and increasing the clock frequency of single-core microprocessors has come to an end (e. g. see, what has happened to the intel Pentium 4 successor project). Instead, multi-core microprocessor chips are emerging from the same vendors. But just more CPUs on the chip is not the way to go for very high performance. This lesson we have learnt from the supercomputing community paying an extremely high price for monstrous installations by having followed the wrong road map for decades. Such fundamental bottlenecks in computer science will necessitate new breakthroughs [1]. Instead of the traditional reductionism we need transdisciplinary approaches, such as heralded by the current revival of Cybernetics, e. g. labelled as Integrated Design & Process Technology [2], or Organic Computing [3]. To reanimate the stalled progress in HPC for a break-through in very high performance computing we need a transdisciplinary approach for bridging the hardware / software chasm, which meanwhile has turned into a configware / software chasm. For much more successful efforts we need a transdisciplinary paradigm shift, over to a new fundamental model, such as available from the Reconfigurable Computing community dealing with configware engineering as a counterpart to software engineering.