High Performance Analytics & Computing Platform – Guidebook

This website is the Guidebook of the Human Brain Project‘s High Performance Analytics & Computing (HPAC) Platform and serves as a collection of all user documentation for the software tools, libraries, frameworks and programming models available as part of the HPAC Platform.

The goal of the HPAC Platform is to build, integrate and operate the hardware and software components of the supercomputing, data and visualisation infrastructure required to:

  • Run large-scale, data intensive, interactive multi-scale brain simulations up to the size of a full human brain.
  • Manage the large amounts of data used and produced by the simulations, and
  • Manage complex workflows comprising concurrent simulation, data analysis and visualisation workloads.

The HPC Platform currently provides supercomputers at four HPC centres for the HBP
community:

JUQUEEN in JUELICH JUQUEEN
Jülich Supercomputing Centre (JSC),
Forschungszentrum Jülich
JURECA at JUELICH JURECA
Jülich Supercomputing Centre (JSC),
Forschungszentrum Jülich
PizDaint at ETHZ-CSCS Piz Daint
Swiss National Supercomputing Centre (CSCS)
MareNostrum at BSC MareNostrum III
Barcelona Supercomputing Centre (BSC)
Marconi MARCONI
Cineca
Pico at CINECA Pico
Cineca

Also the two pilot systems developed in a Pre-Commercial Procurement by Cray and a consortium of IBM-NVIDIA are part of the HPAC Platform:

JULIA (Cray) JURON (IBM-NVIDIA)

Karlsruher Institut für Technologie (KIT) provides the users of the HPC Platform with easy, secure and transparent access to Cloud-based storage. The cloud storage is accessible through an Amazon S3-compatible interface. Rheinisch-Westfälische Technische Hochschule Aachen (RWTH) and École Polytechnique Fédérale de Lausanne (EPFL) provide the HPC Platform users with high fidelity visualisation systems.

The aixCAVE, a five-sided virtual reality environment, is located at RWTH. Both EPFL and RWTH offer tiled display walls as additional high-resolution visualisation systems.