NEW ORLEANS — (BUSINESS WIRE) — November 18, 2014 — International Data Corporation ( IDC) today announced the eighth round of recipients of the HPC Innovation Excellence Award at the SC'14 high performance computing (HPC) conference in New Orleans, Louisiana. Two sets of winners are announced each year, at the November SC conference in the U.S. and the June ISC HPC conference in Germany.
The HPC Innovation Excellence Awards recognize noteworthy achievements by users of high performance computing technologies. The program's goals are to showcase return on investment (ROI) and scientific success stories involving HPC; to help other users better understand the benefits of adopting HPC and justify HPC investments, especially for small and medium-size businesses (SMBs); to demonstrate the value of HPC to funding bodies and politicians; and to expand public support for increased HPC investments.
"IDC research has confirmed that HPC can greatly accelerate innovation and in many cases can generate ROI. The award program aims to collect a large set of success stories across many research disciplines, industries, and application areas," said Earl Joseph, Program Vice President for HPC at IDC. "The winners achieved clear success in applying HPC to improve business ROI, scientific advancement, and/or engineering successes. Many of the achievements will also directly benefit society."
Winners of the prior seven rounds of awards included 41 organizations from the U.S., seven from the UK, four from Italy, three from the People's Republic of China, two each from India and Slovenia, and one each from Australia, Canada, Sweden, South Korea, Switzerland, Germany, France, and Spain.
The new award winners and project leaders announced at SC'14 are as follows (contact IDC for additional details about the projects):
- Argonne National Laboratory, NRG (Netherlands), SCK-CEN (Belgium), TerraPower, and the University of Illinois at Urbana-Champaign: Researchers from Argonne National Laboratory and the University of Illinois at Urbana-Champaign teamed with nuclear reactor designers and research laboratories in the United States and Europe to enable high-fidelity, cost-saving simulations to design the next-generation of nuclear reactors using the computational fluid dynamics code Nek5000. This research will result in multimillion-dollar savings for several companies and nuclear research centers. Project Leads: Paul Fischer, Elia Merzari, Aleks Obabko, and Shashi Aithal, Argonne National Laboratory
- The Center for Pediatric Genomic Medicine at Children's Mercy Hospitals Kansas City was the first genome center in the world to be created inside a children's hospital and one of the first to focus on genome sequencing and analysis for inherited childhood diseases. While most genome centers focus on research, the CPGM develops new clinical tests as a starting point for nextâ�generation medical treatments to improve outcomes in patients at Children's Mercy and around the world. Using the TaGSCAN and STAT-seq applications, Children's Mercy has reduced the overall diagnosis time and substantially helped affected children and their families. Project Lead: Dr. Stephen Kingsmore
- GIS Federal: For the US Army and the intelligence community as a whole, GIS Federal developed an innovative approach to quickly filter, analyze, and visualize in near real time big data streams from hundreds of data providers with a particular emphasis on geospatial data. GIS Federal leveraged the highly parallel compute power of graphical processing units (GPUs) to conduct the data processing. The solution generated multimillion-dollar revenues while saving tens of millions of dollars. Project Leads: Amit Vij and Nima Negahban
- North Carolina State University: Researchers from NCSU conducted innovative research that will allow better prediction of thermal hydraulic behavior for current and future nuclear reactor designs. They analyzed the turbulence anisotropy in single-phase and two-phase bubbly channel flows based on DNS data. These novel simulations will help academia and later industry. Multiphase flow model development for computational fluid dynamics already benefits from high fidelity simulations presented in this work. Project Lead: Igor A. Bolotnov (Department of Nuclear Engineering)
- Nexio simulation is a French SME located in Toulouse and specializing in electromagnetic simulation and studies for marine, space, defense and aeronautics domains applications. Nexio simulation partnered with Bpifrance (the French public bank dedicated to SMEs), Inria, and GENCI to optimize and scale out their current simulation package called CAPITOLE-EM by using large scale HPC resources. While the maximum size of simulation was about 500,000 unknowns in 2011, Nexio was able to simulate 6 millions unknowns thanks to the use of HPC. This major improvement allowed Nexio to win two major contracts with Japanese aerospace companies and to participate in the first PRACE SHAPE call and the Fortissimo project. Project Lead: Pascal De Resseguier
- NASA: The noise generated by civil air transport adversely impacts population centers near major airports. With the expected growth in air travel, community exposure to aircraft noise will increase considerably. To alleviate this problem, the Environmentally Responsible Aviation (ERA) project within NASA's Aeronautics Mission Research Directorate is working to simultaneously reduce aircraft noise, fuel consumption, and engine emissions. High-fidelity simulations are being used to provide an accurate representation of the aerodynamic mechanisms that produce airframe noise (a prominent component of noise during aircraft landing) and to evaluate a suite of novel noise reduction concepts for aircraft flaps and landing gear. Project Lead: Mehdi R. Khorrami
- Central Michigan University researchers used HPC resources to run and visualize a breakthrough simulation involving a long-track EF5 tornado embedded within a supercell. Code was developed to utilize buffered HDF5 output in the CM1 model in order to achieve satisfactory throughput when doing I/O. A plugin was developed to interface VisIt to the CM1 output format. This research adds innovative improvements to the existing simulation workflow, potentially enabling operational use of CM1 models. Project Lead: Leigh Orf
- PayPal engineers developed a platform for realâ�time event analytics using HPC designs on new hardware technology. By converting traditional text data into digital signals through a process of encoding and mapping, the engineers used multi-core digital signal processors in the HP/Texas Instruments Moonshot m800 platform to deliver high performance/lowâ�latency processing with very low power (approx. 11.2GF/watt). A truly revolutionary approach, PayPal's method brings the rich legacy of digital signal processing's capabilities to real-time analytics for the first time. Project Lead: Ryan Quick and Arno Kolster
- The University of Texas MD Anderson Cancer Center, Texas Advance Computing Center (TACC) and Elekta AB: Researchers at MD Anderson Cancer Center in collaboration with TACC and Elekta AB are using detailed Monte Carlo computer simulations of radiation transport to assist in the development of the next generation of radiation therapy cancer treatments, which use a magnetic resonance imaging (MRI) scanner integrated with a radiation therapy unit (MRI-linac unit). The results of the simulations have demonstrated that the response of radiation detectors in the presence of magnetic fields can be predicted and accounted for, enabling researchers to calibrate the new MRI-linac units. This research is expected to lead to the development of new methods and procedures for the use of radiation detectors in the presence of magnetic fields. This will make possible the implementation and safe use of new MRI-linac units. Ultimately, the results of this project will directly contribute to improve treatment outcome of cancer patients. Project Lead: Gabriel O. Sawakuchi
- Researchers at Ohio State University Cancer Comprehensive Care Center developed and implemented bioinformatics and molecular methods to understand what happens to human papillomavirus (HPV) DNA in the "end game" of HPV-positive human cancers. Approximately 15% of all cancers are caused by viruses, including HPV, but the mechanisms by which such viruses cause cancers have remained mostly unknown. Using the new Oakley supercomputer at the Ohio Supercomputer Center, the OSU researchers found that in virtually every HPV-positive cancer cell line that they studied, HPV had integrated into the host genome and was associated with focal genomic instability. They used this insight to develop a new model called the HPV looping model. This research could eventually have a significant impact on how cancer doctors detect and treat different types of virus-associated cancer. Project Leads: Drs. David E. Symer, Keiko Akagi, Jingfeng Li, and Maura L. Gillison, and the Ohio Supercomputer Center
- Tech-X Corporation: To heat magnetically confined plasmas to the millions of degrees needed for fusion reactions, scientists inject megawatts of electromagnetic energy from carefully engineered radiofrequency antennas. The generated electromagnetic waves interact with the plasma in complex ways. Scientists at Tech-X Corporation modeled these interactions using their VSim software at increasing levels of detail, scaling up to 184,000-core, billion-grid-cell simulations on the Titan Cray XK7 at the Oak Ridge Leadership Computing Facility (OLCF). They are now using these simulation results (in collaboration with other members of the SciDAC Center for Simulation of Wave-Plasma Interactions) to develop and refine predictive models for wave-plasma interactions in the reactor core and edge. Such computations enable the identification of more efficient operational regimes for existing magnetic fusion experiments, and provide predictive capabilities for future experimental devices. This work was funded by DOE grant DE-SC0009501, with OLCF computing resources provided by an ALCC award sponsored by DOE's Office of Science, Contract No. DE-AC05-00OR22725. The VSim team provided the VSim computational application. Project Leads: Thomas G. Jenkins and David N. Smithe