Navigation and service

JULIA and JURON: Pilot Systems for Interactive Supercomputer

The Jülich Supercomputing Centre began operation of two new pilot systems for an interactive supercomputer in September 2016. The systems – JULIA, created by Cray, and JURON from IBM and NVIDIA – are specifically designed for applications in the neurosciences. The installation and test phase are part of a special procurement for the European Human Brain Project.

The objective is a computer with a peak performance of initially 50 petaflop/s and a memory capacity of 20 petabytes, with which large-scale brain simulations can be interactively visualized and controlled.

Two consortia have installed their respective proposed solutions at the Jülich Supercomputing Centre (JSC), where Jülich scientists and their colleagues from the HBP now test how powerful the two systems are and how well they fulfil the desired functions.

JULIA and JURON comprise two racks each. The pilots will share a storage system with two racks.

JULIA (Cray)

Cray introduced new processor and network technology from Intel in the JULIA supercomputer, whose name is derived from JUelich and GLIA, a type of cell in the nervous system. In order to offer more memory in the system, special DataWarp nodes were integrated, which have a lot of non-volatile memory at their disposal. The results of both approaches are to be reviewed in early 2017.

JURON (IBM/NVIDIA)

For JURON (the name is derived from JUelich and NeuRON), IBM and NVIDIA make use of completely novel Tesla graphics processing units of the Pascal generation, which are capable of particularly rapid communication with each other and with IBM's POWER8 processor on the basis of the NVLink technology.

For their concepts, the two consortia are making use of fast computing technologies, not all of which are yet commercially available.

6 Racks: 2 von JULIA, 2 des gemeinsam genutzten Speichersystems, 2 von JURONThe two new pilot systems for an interactive supercomputer, which are installed at Jülich Supercomputing Centre: JULIA, created by Cray, and JURON from IBM and NVIDIA. The central racks represent the storage system both pilot systems can access. Both systems are specifically designed for applications in the neurosciences. The installation and test phase are part of a special procurement for the European Human Brain Project.
Copyright: Forschungszentrum Jülich

Time-lapse video of the installation

noPlaybackVideo

DownloadVideo

Functional tests

Jülich scientists and their colleagues from the HBP will now test how powerful the two systems are and how well they fulfil the desired functions. For this purpose, they are using simulation software such as NEST, which Jülich scientists headed by Prof. Markus Diesmann from the Institute of Neuroscience and Medicine (INM-6) are constantly further developing together with neuroscientists from all over the world. NEST makes it possible to simulate neuronal networks in the brain. It is planned that the resulting data can in future be directly analysed and compared to experimental data using the novel supercomputers. In order to achieve this, an international team of scientists headed by Prof. Sonja Grün (INM-6) together with JSC's Simulation Laboratory Neuroscience are further developing the "Elephant" software. Methods for data and image analysis developed by Jülich researchers headed by Prof. Katrin Amunts from the Institute of Neuroscience and Medicine (INM-1) are also employed. Among other aspects, three-dimensional models of the human brain such as the BigBrain project are to be realized, as well as maps recording the connections between different regions of the brain.

Key element in the neurosciences of the future

Being able to interactively operate supercomputers for such applications is viewed as a key element for future neuroscience. So far, supercomputers work on tasks largely autonomously. The aim for the future is for scientists to be able to interact with their jobs and to control them interactively. The special challenge is that this requires data to be rapidly analysed and visualized in parallel to the main application. This means that the computers of the future must handle even more data than the enormous amounts that already arise today.

Pre-Commercial Procurement (PCP)

Both systems serve to evaluate research and development work as part of Pre-Commercial Procurement (PCP). The Human Brain Project chose this special type of call for tenders funded by the European Commission in order to find suitable technology solutions for the high-performance computers of the future designed specially for neuroscience. The procurement process began in April 2014 and is planned to be concluded in 2017. Its aim is to support the development of technologies to be integrated in future products. Forschungszentrum Jülich hopes that in this way, suitable solutions for the development of the HBP High Performance Analytics & Computing Platform will become available on the market.

Further information on the pilot systems:

JULIA pilot system (Cray)

JURON pilot system (IBM/NVIDIA)

 Pilot Systems For the Human Brain Project (PDF, 331 kB)

Further Information on the Human Brain Project

Human Brain Project Website