SuperMUC-NG will not only significantly improve the compute power, but also enable the handling of tremendous amounts of data (Big Data) accumulated in today's experiments and simulations. A further aim is to give users the flexibility to deploy their own software and visualisation environments for analyzing the data generated by simulations and sharing the results. For better integration with modern concepts of handling and visualisation of huge amounts of data, SuperMUC-NG will be linked to separately operated Cloud components delivered with the system.
The new supercomputer at LRZ will be equipped with more than 6.400 Lenovo ThinkSystem SD650 DWC compute nodes based on the Intel Xeon Scalable processor. They will be connected by Intel's Omni-Path high-speed interconnect, using a "fat tree" topology. The overall more than 300,000 compute cores will deliver 26.7 Pflop/s (26,700,000,000,000,000 floating point operations per second) peak performance. 700 TByte main memory and more than 70 PByte disk storage will be available. Using next-generation interconnects and providing greater storage capabilities means that LRZ will be better-suited than ever before to address the increasingly difficult challenges of data management.
The new system will also include the integration of Lenovo Intelligent Computing Orchestrator (LiCO), a powerful management suite with an intuitive GUI that helps accelerate development of HPC and AI applications, as well as Cloud-based components to empower LRZ researchers with the freedom to virtualize, process the vast amount of data sets and expediently share results with colleagues.
"We are happy to contribute an essential part to this important project, thereby supporting LRZ's service efforts. Handling huge amounts of data is vital", stated Hannes Schwaderer, Head of Enterprise Sales Intel Deutschland GmbH. "Processing these data requires immense computational power. Intel's state-of-the-art processor architecture provides an important basis and ensures that all demands for high performance computing can be optimally fulfilled."
Just like SuperMUC, SuperMUC-NG will be cooled using warm water. Lenovo, the system integrator, has developed a cooling concept that will further reduce power consumption and reuse the waste heat of the supercomputer to generate cold water. "As a leading technology provider for high-performance data centres we focus our innovations on performance, reliability and sustainability", stated Scott Tease, Executive Director, HPC and AI, Lenovo Data Center Group. "All of this contributes to the cooperation with Leibniz Supercomputing Centre and Intel regarding SuperMUC-NG."
"Lenovo is committed to providing research institutions like LRZ with not only sheer computing power, but a true, end-to-end solution that can help effectively and efficiently solve critical humanitarian challenges. We're pleased to be working on this next-generation project in partnership with Intel", stated Scott Tease. "The new SuperMUC-NG installation will provide LRZ with greater compute power in a smaller data centre footprint with drastically reduced energy usage through innovative water-cooling technology, offering researchers a comprehensive supercomputing solution that packs more performance than ever to accelerate critical research projects."
In combination with Lenovo Energy Aware Run-Time (EAR) software, a technology that dynamically controls system infrastructure power while applications are still running, Lenovo's comprehensive water-cooling technology delivers 45 percent greater electricity savings to LRZ as compared to a similar, standard air-cooled system. Together, these energy efficiency innovations will help further reduce the research centre's carbon footprint and total cost of ownership.
Up to now, cosmology and astrophysics, solid-state physics and fluid dynamics have been the classical domains of supercomputing. During the last years other scientific domains have entered the field, mainly projects from life sciences as well as environmental and disaster research.
Just two recent examples where SuperMUC has enabled scientific breakthrough: An international team studying personalized medicine figured out which treatment for breast cancer will be efficacious for which specific patient. And a simulation of the Sumatra-Andaman Earthquake causing the tsunami in 2004 was awarded "Best Paper" at the Supercomputing Conference in Denver, Colorado, USA.
A team led by Technical University of Munich Professor Dr. Michael Bader and Ludwig-Maximilians-Universitaet Munich researcher Dr. Alice-Agnes Gabriel used the current-generation SuperMUC to create the largest, longest ever multiphysics simulation of an earthquake and its resulting tsunami. Dr. Bader indicated that next-generation machines would help his team run many more iterations of its simulations. By testing its models with larger numbers of input parameters, he anticipates being able to achieve a better understanding of how earthquakes and tsunamis develop. Ultimately, this could lead to real-time solutions for mitigating their risks.
"Currently, we've been doing one individual simulation at a time, trying to accurately guess the starting configurations for things like the initial stresses and forces, but all of these are uncertain", he stated. "We would like to run our simulation with many different settings to see how slight changes in the fault system or other factors can impact the study, but such large parameter studies would require yet another layer of performance from a machine."
Researchers also use the supercomputer to predict environmental hazards: Together with their colleagues from Canada, scientists at Ludwig-Maximilians-Universitaet and Technical University Munich and LRZ performed comprehensive simulations of the effects of global warming on extreme rainfalls and floodings in Bavaria.
Dieter Kranzlmueller, Chairman of the Board of LRZ, sees the Supercomputing Centre well equipped for the future: "The new supercomputer SuperMUC-NG will provide more compute power for scientists, but also will require more expertise. Researchers will be able to tackle problems that are more complex. The LRZ experts assist them, providing an interface between the scientific community and computer science. We are well prepared to support scientists in achieving the next level of supercomputing. As part of the project we will again extend our user support team."
Upon its completion in late 2018, the new supercomputer will support LRZ in its groundbreaking research across a variety of complex scientific disciplines, such as astrophysics, fluid dynamics and life sciences, by offering highly available, secure and energy-efficient high-performance computing (HPC) services that leverage industry-leading technology optimized to address the a broad range of scientific computing applications. The LRZ installation will also feature the 20-millionth server shipped by Lenovo, a significant milestone in the company's data centre history.
The funding of SuperMUC-NG is shared to equal parts by the federal government of Germany and by the Free State of Bavaria through a strategic plan of the Gauss Centre for Supercomputing (GCS). GCS assumed half the cost with the two governments matching the other half. The total cost of the project sums up to 96 million euro for 6 years including electricity, maintenance and personnel. Bavaria's Minister of Science Dr. Ludwig Spaenle stated: "Excellent research and development need excellent working conditions. With the next supercomputer SuperMUC-NG, we meet these demands and establish the prerequisites for continuation of state-of-the-art scientific research in this field in Bavaria."
Once operational, the LRZ SuperMUC-NG system is expected to place on the industry-wide TOP500 list.