header news2

This website uses cookies to manage authentication, navigation, and other functions. By using our website, you agree that we can place these types of cookies on your device.

View Privacy Policy

Last week at the ICT 2015 Conference in Lisbon, Prof Thomas Lippert was invited as speaker to the panel on “Multidisciplinary for Future Technology”. The Head of Jülich Supercomputing Centre and Chief Scientific Coordinator of the DEEP and DEEP-ER projects underlined in which ways the multi-disciplinary nature of the projects helps to unleash breakthroughs in computer science and explained the new compute paradigm the projects have devised.

A new computing paradigm

Looking at the challenges future supercomputing architectures have to overcome, the question arises if the leap to Exascale computing can only be achieved by special-purpose next-gen systems with diverging properties. The novel concepts the DEEP and DEEP-ER projects put forward challenge this idea and contrast it with a modular approach to future supercomputers: The system features different types of processors that facilitate different concurrency levels. It also employs a fine-grained memory hierarchy with new technologies like non-volatile and network attached memory and a massively parallel file system to just mention a few building blocks. All these components can be seen as pool of resources HPC application scientists can choose from. Designing this sophisticated architecture was only possible through a continuous co-design effort and vivid exchange between hardware, system software, and application developers. All in all, this novel architecture allows for diverging properties within a single system and its modular concept could also allow for integrating completely new technologies like bio or quantum computing once available. 

Starting the experiment

The DEEP and DEEP-ER projects are set out to prove the feasibility of this modular approach to supercomputing. The first DEEP prototype is now up and running at Jülich Supercomputing Centre provisioning a peak performance of 500 TFlop/s and subjects the idea for modular supercomputing to actual experiments.

Broadcast available

But see and listen for yourself – Thomas Lippert’s presentation is available as video via the ICT 2015 conference website: http://tinyurl.com/ofp34ke