30/03/2009 The Race of the Century Universities, research labs, corporations and governments from around the world are lining up for the race of the century It s a race to solve many of the most complex problems mankind has ever considered Being fast not only determines who gets knowledgeable or wealthy and who doesn t, but will be important for our planet and our survival European View on Supercomputing Prof. Thomas Lippert, Jülich Supercomputing Centre Launch of the Flemish Supercomputer Center (VSC), Leuven, March 23, 2009 23.3.2009 Supercomputing Drives Applied Science Thomas Lippert 2 Supercomputing Drives Basic Sciences New Physics Nuclear Physics Plasma Physics Geophysics Environment Weather/Climatology Pollution / Ozone Hole 23.3.2009 Ageing Society Medicine Biology Thomas Lippert Materials Spintronics Nano--Science Nano Energy Plasma Physics Fuel Cells 3 23.3.2009 Astrophysics Astrophysik Kosmologie Cosmology Particle Physics Thomas Lippert Structure of Matter Solid State Physics Chemistry 4 1
Supercomputing Drives Engineering and Business Competitiveness Reducing design costs by virtual prototyping: faster time to market Allowing investigations where economics or ethics preclude experimentation FROM JÜLICH TO EUROPE imperative of supercomputing 5 6 Forschungszentrum Jülich (FZJ) Jülich in Brief Largest civilian research centre in Europe 4.300 staff members, 1.200 scientists 9 Departments (Institutes) Simulation has become a key activity 7 8 2
Jülich Supercomputing Centre (JSC) Supercomputers @ JSC 2004 IBM Power 4+ JUMP, 9 TFlop/s 2005/6 IBM Blue Gene/L JUBL, 45 TFlop/s 2007/8 IBM Power 6 JUMP, 9 TFlop/s File Server GPFS IBM Blue Gene/P JUGENE, 223 TFlop/s 2009 Intel Nehalem Clusters JUROPA 200 TFlop/s HPC-FF 100 TFlop/s File Server GPFS, Lustre IBM Blue Gene/P JUGENE, 1 PFlop/s General-Purpose Highly-Scalable 9 10 Levels of User Support and Training User Support Training German Scientific Case Scientific Field (numbers in TeraFlop/s) 2005-2007 2007-2009 2010 Climate and Earth System Research 20 50-100 >500 Geophysics 1 10-100 >1000 Nanostructure Physics 1 10-50 >200 Solid-State Physics 1 50-100 >1000 Computational Fluid Dynamics 2.5 25-100 >1000 Astrophysics 10 50-100 >500 Elementary Particle Physics and Physics of Hadrons and Nuclei 30 100 >1000 Materials Science 10 50-100 >500 Theoretical Chemistry 3 25-125 >300 Soft Matter 3 30 >200 Biophysics and Bioinformatics 3 15-80 >1000 Plasma Physics 10 50 >500 A. Bode, W. Hillebrandt, and Th. Lippert: German Scientific Case for the BMBF, 8/2005 11 12 3
Gauss Centre for Supercomputing HPC Pyramid in Germany European HPC Centre Gauß Centre for Supercomputing National HPC Centre 3 Garching, Jülich, Stuttgart Germany s Tier-0/1 Supercomputing Complex Association of Jülich, Garching and Stuttgart A single joint scientific governance Germany s representative in PRACE More information: http://www.gauss-centre.de Topical HPC Centre, Centre with regional tasks HPC Server ~ 10 ~ 100 Aachen, DKRZ, Dresden, DWD, Erlangen, G-CSC Frankfurt, HLRN Gauß(Hannover, Alliance Berlin), Karlsruhe, MPG/RZG, Paderborn University/Institute 13 14 European Users of JSC Facilities CSC DEISA: Distributed Supercomputing Infrastructure Edinburgh Glasgow Roskilde Gdansk I3HP: Hadron Physics SARA ECMWF IDRIS HLRS Warsaw Prague Cracow LRZ Brno RZG Bratislava Vienna Budapest NIC Initiative towards new EU Member States Coimbra BSC CINECA Rome Zagreb Athens Nicosia Other Collaborations Towards the High-End HPC Service for European Science 15 4
HET: The Scientific Case Weather, Climatology, Earth Science degree of warming, scenarios for our future climate. understand and predict ocean properties and variations weather and flood events Astrophysics, Elementary particle physics, Plasma physics systems, structures which span a large range of different length and time scales quantum field theories like QCD LHC, FAIR ITER Material Science, Chemistry, Nanoscience understanding complex materials, complex chemistry, nanoscience the determination of electronic and transport properties Life Science system biology, chromatin dynamics, large scale protein dynamics, protein association and aggregation, supramolecular systems, medicine Engineering complex helicopter simulation, biomedical flows, gas turbines and internal combustion engines, forest fires, green aircraft 17 Computational science infrastructure in Europe The European Roadmap for Research Infrastructures is the first comprehensive definition at the European level Research Infrastructures are one of the crucial pillars of the European Research Area A European HPC service:: strategic competitiveness attractive for researchers supporting industrial development 18 ESFRI Vision for a European HPC service 6 years technolgy advantage with a #1 system European HPC-facilities at top of an HPC provisioning pyramid Tier-0: 3-5 European Centres Tier-1: National Centres Tier-2: Regional/University Centres Creation of a European HPC ecosystem HPC service providers on all tiers Grid Infrastructures Scientific and industrial communities The European HPC industry tier-0 tier-1 tier-2 PRACE DEISA Renewal every 2-3 years Construction cost 200 400 Mio. Annual running cost 100 200 Mio. 19 Top 500 list 11/07 6 years after 12 years the power is on laptop 20 5
Steps and Achievements PRACE Project Consortium tier 1 Production of the HPC part of the ESFRI Roadmap; Creation of a vision, involving 15 European countries Signature of the MoU Bringing scientists together Creation of the Scientific Case Submission of an FP7 project proposal Approval of the project Project start Principal Partners tier 0 HPCEUR HET PRACE Initiative 2004 2005 2006 2007 2008 General Partners New Partners - since May 2008 - of the PRACE Initiative 21 22 PRACE Project Facts Partners 16 Legal Entities,14 countries Project duration January 2008 December 2009 Project budget 20 M, EC funding: 10 M Coordinator JSC-FZ Jülich PRACE is funded in part by the EC under the FP7 Capacities programme grant agreement INFSO-RI-211528 Main Goal Create and implement a High Performance Computing (HPC) service as a persistent and sustainable European Research Infrastructure, hosted by 3-5 Tier-0 centres 23 24 6
Tasks HPC Services for European Industry? Prepare contracts Usage of HPC technology in industry 6-8 Establish the PRACE Research Infrastructure Single Legal Entity in 2010 Governance, funding, procurement, and usage strategies. Technical work Preparation of operation of Tier-0 systems in 2009/2010 Deployment and benchmarking of Petaflop/s prototypes Peta-scaling of applications years behind technology frontier The USA (and Japan) boost competitiveness of their industries by US: free-of-charge access through INCITE program Japan: initially free of charge PRACE is striving towards similar models 25 26 First Industry Seminar attendees Revitalizing HPC Industry in Europe Access to HPC-technology is strategic for EU PRACE will foster European developments by translating user requirements to architectural specifications for future multi-petascale HPC systems Cooperating with consortia of industrial and academic stakeholders to develop future components and systems Enforce cooperation with HPC-Industry based in US, Japan, China... PRACE Prototypes and development of prototypes and components 27 28 7
PRACE All-Hands meeting February 2009 Strategic Approach Experience from other European Research Infrastructures Negotiations on political level of all critical questions Balancing national interests and European vision PRACE approach Continue intense consultations on national level New members of the initiative are welcome! Initiative members address important issues beyond the scope of the project through new FP7 proposals (CSAs, ERA-NETs) 29 30 Mitglied der Helmholtz-Gemeinschaft Ozone Hole over Germany (Prof. M. Riese, FZJ) Supercomputer as Crystal Ball Environment 31 32 8
Mitglied der Helmholtz-Gemeinschaft Simulation of Blood Flow in a Ventricular Assist Device (Prof. Marek Behr, RWTH Aachen) Virtual Prototyping Health 33 34 Mitglied der Helmholtz-Gemeinschaft Supercomputer as virtual Magnetic Surfaces (Prof. S. Blügel, FZJ) Spin forces on surfaces know about left and right S. Blügel et al., Nature 447, 441 446 (10 May 2007) (Jülich) Upper configuration can exist Nano-Lab New Materials Mirrored configuration is unstable 35 36 9
Mitglied der Helmholtz-Gemeinschaft The Origin of Life (Prof. D. Marx, RU-Bochum) Activated amino acids form peptides outside cells Supercomputer as virtual Test Tube Life 37 38 Mitglied der Helmholtz-Gemeinschaft Understanding the World of the Quarks Supercomputer as Femto-Scope Particle Physics 39 Z. Fodor, BU-Wuppertal, S. Krieg, FZJ, SCIENCE 40 10
Z. Fodor, S. Krieg, Th. Lippert et al., SCIENCE, 2008 Validating Quantum Chromodynamics Supercomputing: Looking Beyond Experimental Insight 41 42 11