Navigation and service

HPC Technology

Giga, tera, peta: in a period of just 20 years the performance of the world’s fastest computers has increased by a factor of one million. Although in the past this was due above all to an increase in the processors’ clock frequency, today the performance of top systems like JUGENE is based on a gigantic number of processors operating in parallel. The next target the developers have in their sights in research centres and industry is a performance of one exaflop per second – a one with 18 zeros, one quintillion. Today we know that this goal requires the readiness to develop and accept whole new paradigms in many respects. The many millions of processor cores must be synchronized, the reliability of the components must be guaranteed, and new storage technologies are required.

A quite decisive part will be played by reducing energy consumption. As a partner of the Collaborative Research Centre Transregio “Hadron Physics”, headed by the University of Regensburg, we have shown how this can be done. Our joint computer development QPACE is currently regarded as the most energy-efficient computer in the world. On the one hand, we shall continue along this path to the exaflop/s computer by our foundation of the “Exascale Innovation Centre” together with IBM. On the other hand, we are working with Intel and ParTec on developing the cluster computers of the future at the Jülich “ExaCluster Laboratory” (ECL).