The true power of supercomputing is in the process of taking a massive leap forward. A collaboration between IBM and GENCI over 18 months will hopefully bring exascaling computing to the current world of supercomputing.
On Friday, a formal announcement of IBM’s collaboration with a French high performance computing agency went out. This collaboration is aimed at speeding the process of developing and refining exascaling computing.
Exascale computing is the term used to describe computers having the ability to process up to speeds of one exaflop in one second.
To put that into laymans terms, with the development of this computing power, the current fastest processing machine would be one-third slower than a computer able to perform exascalating computing. The current fastest processing is around 10 to 33 petaflops, or about 10 to 33 million billion calculations per second.
As part of this collaboration, which is expected to last around 18 months, the companies will use computing technology from the OpenPOWER ecosystem. This uses a different type of computer technology that is widely supported by over 140 OpenPOWER Foundation members and thousands of developers across the globe.
Part of the expectations for this system under development from these two companies are a computing processing power that can exceed 100 petaflops.
For this to be achieved, GENCI will focus on the impact and requirements of the OpenPOWER architecture on any scientific applications in order to better understand the requirements to reach the exascaling computing.
The CTO of IBM France, Michel Teyssedre, said, “The work we are doing with GENCI-bringing together some of the best minds in science and information technology-is a collaborative effort on a grand scale involving not just GENCI and IBM, but thousands of developers contributing to the rapidly expanding OpenPOWER ecosystem worldwide.”