Science applications requiring exascale compute and data capabilities Jack Wells Director of Science Oak Ridge Leadership Computing Facility Oak Ridge National Laboratory SIAM EX14 Workshop Chicago 6 July 2014 ORNL is managed by UT-Battelle for the US Department of Energy
Outline • U.S. DOE Leadership Computing Program – Breakthrough Science Across a Broad Range of Disciplines • Science opportunities over next decade – Fusion Energy, Biomass to Biofuels, Solar Energy, Nuclear Energy • Industry also has big problems and demanding requirements • CORAL Procurement: Mission need for pre- exascale capability system in 2018 • Exascale challenges remain with us 2
Mission need for Leadership Computing Leadership computing capability is required for scientists to tackle the high-resolution, multi-scale/multi-physics simulations of greatest interest and impact to both science and the nation. Leadership Computing capability is typically 10-100X greater than other computational centers. Leadership Computing research is mission critical to inform policy decisions and advance innovation in far reaching topics such as: • energy assurance “We will respond to the • ecological sustainability threat of climate change, • scientific discovery knowing that the failure to do so would betray our • global security children and future generations.” – President Obama 1/21/2013 3
What is the Leadership Computing Facility (LCF)? • Collaborative DOE Office of Science • Highly competitive user allocation program at ORNL and ANL programs (INCITE, ALCC). • Mission: Provide the computational • Projects receive 10x to 100x more and data resources required to solve resource than at other generally the most challenging problems. available centers. • 2-centers/2-architectures to address • LCF centers partner with users to diverse and growing computational enable science & engineering needs of the scientific community breakthroughs (Liaisons, Catalysts). 4
ORNL has increased system performance by 1,000 times 2004-2010 Hardware scaled from single- Scaling applications and system core through dual-core to software was the biggest challenge quad-core and dual-socket , 12-core SMP nodes Cray XT5 Systems 12-core, dual-socket SMP Cray XT4 Cray XT4 2.3 PF Quad-Core Dual-Core Cray XT3 Cray XT3 263 TF 119 TF Dual-Core Single-core Cray X1 54 TF 26 TF 3 TF 2007 2005 2006 2008 2009 5
Science breakthroughs at the OLCF: SELECTED science and engineering advances over the period 2003 - 2013 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 MD simulations show selectivity Researchers solved the 2D Hubbard model filter of a trans-membrane ion and presented evidence that it predicts channel is sterically locked HTSC behavior, Phys. Rev. Lett ( 2005) open by hidden water 105 citations, 3/2014 molecules, Nature (2013) First-Principles Flame Simulation Provides Crucial Information to Guide Design of Fuel-Efficient Clean Calculation of the number of bound Engines, Proc. Combust. Insti. (2007) nuclei in nature, Nature (2012), 36 78 citations, 3/2014 citations, 3/2014 , 36 citations, 3/2014 Largest simulation of a galaxy’s worth of Global Warming preceded by increasing dark matter, showed for the first time the carbon dioxide concentrations during the fractal-like appearance of dark matter last deglaciation, Nature (2012). substructures, Nature (2008) 64 citations, 3/2014 326 citations, 3/201 4 World’s first continuous simulation of Demonstrated that three-body forces Astrophysicists discover 21,000 years of Earth’s climate are necessary to describe the supernova shock-wave instability, history, Science (2009) long lifetime of 14 C Astrophys. J. (2003) 116 citations Phys. Rev. Lett. (2011) 254 citations, 3/2014 28 citations, 3/2014 Biomass as a viable, sustainable feedstock for hydrogen production for fuel cells, Nano Letters (2011) J. Phys. Chem. Lett. (2010) 71 & 74 citations, respectively 6
High-impact science across a broad range of disciplines For example in 2013: Molecular Biology Paleoclimate Science “Recovery from slow “Northern Hemisphere forcing inactivation in K1 channels is of Southern Hemisphere controlled by water molecules” climate during the last Jared Ostmeyer, et al. (U. deglaciation,” Chicago) Nature , Sept. (2013) Feng He (UW Madison), et al ., Nature , February (2013) Superconductivity Molecular Biology “ Doping dependence of spin “ A phenylalanine rotameric excitations and correlations switch for signal-state control in with high-temperature super- bacterial chemoreceptors” conductivity in iron pnictides,“ D. Ortega (UTK), Meng Wang(IOP CAS Beijing), Nature Communications Nature Communications . December (2013) December (2013) Complex Oxide Materials Polymer Science “Atomically resolved “Self-Organized and Cu- spectroscopic studyof Sr2IrO4: Coordinated Surface Linear Experiment and theory,” Qing Polymerization” Li (ORNL), E.G. Eguiluz (UTK) Qing Li, B. Sumpter (ORNL), Nature Scientific Reports . Nature Scientific Reports . October (2013) July (2013) 7
Science opportunities over the next decade Application requirements process has been guiding paths forward • Science benefits and impact of future systems are examined on an ongoing basis. • Baseline plans are developed in consultation with leading domain scientists. • Detailed performance analyses are conducted for a subset of applications to understand architectural bottlenecks. • Leadership Computing Facility has been actively engaged in community assessments of future computational needs and solutions. The image cannot be displayed. Your computer may not have enough memory to open the image, or the image may have been corrupted. Restart your computer, and then open the file again. If the red x still appears, you may have to delete the image and then insert it again. 8
Our Science requires that we advance computational capability 1000x over the next decade. Mission: Providing world-class Vision: Deliver transforming computational resources and specialized discoveries in climate, materials, services for the most computationally biology, energy technologies, etc intensive global challenges Roadmap to Exascale OLCF-5: 1 EF 20 MW OLCF-4: 100-250 PF Titan 27 PF 4000 TB memory What are the Jaguar 2.3 PF 600 TB DRAM > 20MW Challenges? 362 TB DRAM Hybrid GPU/CPU 6 day resilience 2010 2012 2017 2022 9
Science challenges for LCF in next decade Combustion Science Climate Change Science Increase efficiency by Understand the dynamic 25%-50% and lower ecological and chemical emissions from internal evolution of the climate combustion engines using system with uncertainty advanced fuels and low- quantification of impacts. temperature combustion. Fusion Energy Biomass to Biofuels Develop predictive Enhance the understanding understanding of plasma and production of biofuels for properties, dynamics, and transportation and other bio- interactions with products from biomass. surrounding materials. Nuclear Energy: Solar Energy For existing reactors, provide Improve photovoltaic safe, increased fuel utilization, efficiency and lower power upgrades, and reactor cost for organic and lifetime extensions. Design new, inorganic materials. safe, cost-effective reactors. . 10
Fusion Energy/ITER Key science challenges: Effectively model and control the flow of plasma and energy in a fusion reactor, scaling up to ITER-size. Develop predictive understanding of plasma properties, dynamics, and interactions with surrounding materials. Mitigate plasma disruptions. A global particle-in-cell simulation to show core turbulence in a tokamak. Image S. Ethier, PPPL Science enabled by LCF Capabilities 2013-2018 2018-2023 • Perform high-fidelity simulation of edge plasma • Perform integrated first-principles simulation turbulent transport in tokamak from first- including the critical multiscale processes to principles to address DIII-D and JET-scale study fusion-reacting plasmas in realistic plasmas with a goal of understanding high- magnetic confinement geometries. confinement physics. • Produce an experimentally validated • Increase simulation of tokamak edge plasma to simulation capability for ITER to design ITER scale. Coupled simulations of plasma edge DEMO, the ITER follow-on facility to solve the with core and chamber wall interactions. Control engineering issues necessary for electricity edge-localized modes and other destructive production with fusion plasmas. mechanisms. 11
XGC1 on Titan simulates turbulent transport in plasma edge for whole tokamak from first-principles Courtesy: CS Chang XGC1 Gyrokinetic simulation of “blobby” The resulting heat load footprint from edge turbulence in a DIII-D H-mode XGC1 on divertor plate, mapped back plasma, together with background and to outboard midplane. neutral particle dynamics. Achieved clarification of some assumptions/findings from reduced models, together with some new discoveries 12
Recommend
More recommend
