RESEARCH GIANT IBM will develop an exascale supercomputing cluster as part of a five year project with the Netherlands Institute for Radio Astronomy (ASTRON).
IBM has announced it entered a €33m research project with ASTRON to build a low-power exascale supercomputer cluster that will sift through data collected by the international Square Kilometre Array (SKA). The SKA is expected to produce exabytes of data every day peering back through time to find the origin of the universe, which until recently many hacks believed to be at the bottom of a beer glass.
Ton Engbersen of IBM Research in Zurich said, "If you take the current global daily Internet traffic and multiply it by two, you are in the range of the data set that the Square Kilometre Array radio telescope will be collecting every day. This is Big Data Analytics to the extreme. With DOME we will embark on one of the most data intensive science projects ever planned, which will eventually have much broader applications beyond radio astronomy research."
The SKA itself isn't expected to come online until 2024, but building an exascale supercomputer cluster isn't a turnkey research project either. Although IBM bandied around data storage numbers the real challenge is not storage but power efficiency.
Intel has said exascale computing is one of its biggest challenges, and admitted that present designs need to be modified significantly for exascale clusters to become possible. While it is theoretically possible to simply increase the number of nodes based on current technology, cooling the nodes becomes prohibitive, meaning chip and interconnect designers have to create a low power design if they intend to break the exascale barrier. µ
Sign up for INQbot – a weekly roundup of the best from the INQ