CLOUD

International Data Corporation (IDC) announced the fourth round of recipients of the HPC Innovation Excellence Award at the SC12 supercomputer industry conference in Salt Lake City, Utah. Prior winners were announced at the ISC'11, SC11, and ISC’12 supercomputing conferences.

The HPC Innovation Excellence Award recognizes noteworthy achievements by users of high performance computing (HPC) technologies. The program's main 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 shown that HPC can greatly improve ROI and scientific advancement. The award program aims to collect a large set of success stories across many research disciplines, industries and application areas," said Chirag Dekate, research manager, High Performance Computing at IDC. "The winners achieved clear success in applying HPC to greatly improve business ROI, scientific advancement, and/or engineering successes. Many of the achievements also directly benefit society."

Winners of the first three rounds of awards, announced in 2011 and early 2012, included 13 organizations from the U.S., three from the People's Republic of China, one from Canada, and one from India.

The new award winners and project leaders announced at SC12 are as follows (contact IDC for additional details about the projects):

• Computational Research Laboratories (INDIA). Engineers at Computational Research Laboratories (CRL), a subsidiary of Tata Consultancy Services, developed an HPC-driven computational fluid dynamics (CFD) procedure aimed at adding realism to real-world wind tunnel data. This innovation could become crucial for the aerospace industry and others using similar technology to improve their product designs. The CFD developments by CRL-TCS are targeted at increasing the accuracy of predictions of maximum lift for aircraft and optimizing the utilization of wind tunnel testing. Improving maximum lift calculations could enable designers to use less material in aircraft, resulting in improved fuel efficiency and lower fuel costs. Improvements for wind tunnel testing alone have resulted in up to 20% reduction in flap prediction tests and estimated savings of at least $2 million per product design. The CFD methodologies could potentially be applied to other problems, resulting in more savings. The investments to enable the CRL-TCS results, including the system cost, were between $1 million and $2 million. The simulations were conducted on the Hewlett Packard EKA system. Project Leader: Kishore Nikam and Anutosh Moitra
• St. Vincent’s Institute of Medical Research (AUSTRALIA.). A cross-organizational team comprising researchers from St. Vincent’s Institute of Medical Research, Victorian Infectious Disease Research Laboratories, IBM Research Collaboratory for Life Sciences – Melbourne, and Victorian Life Sciences Computation Initiative developed a method to simulate the 3D atomic motion of the complete human rhinovirus on Australia’s fastest supercomputer, paving the way for new drug development. This research is the first time that the atomic motion of a complete human rhinovirus has been simulated on a computer. Understanding how anti-viral drugs work on rhinoviruses and related viruses can potentially speed up the development of new treatments, and could produce savings in development costs. The research has the potential to produce savings in drug discovery and pre-clinical development of up to $1,000,000 per year. The research was conducted on an IBM BlueGene Q system and the storage infrastructure was developed by DDN. Project Leader: Professor Michael Parker, with Jason Roberts, Matthew Downton, and Michael Kuiper
• Cycle Computing (U.S.). Cycle Computing provisioned an over 50,000-core utility supercomputer in the Amazon Web Services (AWS) cloud for Schrödinger and Nimbus Discovery to accelerate lead identification via virtual screening. Schrödinger ran a virtual screen of 21 million compounds against a protein target. The run cost under $4,900 per hour at peak and required no upfront capital. The cloud-based run time per job averaged 11 minutes and the total work completed topped 100,000 hours. Project leader: Jason Stowe
• Intelligent Light (U.S.). The CFD workflow and software developed by Intelligent light is benefiting small and medium size manufacturer Zipp Speed Weaponry. A direct consequence of the work was a rapid change in design direction to bring a revolutionary new racing wheel, “Zipp Firecrest,” to market. HPC and a highly productive, automated CFD workflow enabled Zipp to create a sustainable competitive advantage by helping them effectively design and functionalize aerodynamic racing wheels for cyclists. The hardware system was powered by an on-demand cloud from Dell R Systems. This project involved an aggregate investment of $130,000. Project leader: Roger R. Rintala
• PayPal (U.S.). PayPal, an eBay company, leverages HPC technologies to provide real-time stream analytics for behavioral and situational modeling related to online fraud detection. The addition of HPC resources from SGI, Hewlett Packard, Mellanox and QLogic (Intel) has enabled PayPal to advance to real-time fraud detection. Together, the system handles more than 4 billion new records daily. Conservatively estimated, the revenue savings generated by applying the HPC infrastructure for fraud detection translates to over $710 million. Project leader: Arno Kolster and Ryan Quick
• Ramgen Power Systems LLC (U.S.). Ramgen Power Systems is a small business in Washington State that is developing a novel gas compressor system based on shock wave compressor technology. The system will be used in supersonic flight applications. Use of the Cray "Jaguar" supercomputer at Oak Ridge National Laboratory cut down R&D costs by over $4 million. Using Jaguar, staff of the Oak Ridge Leadership Computing Facility were able to give these applications a 100-fold speedup in time to solution and two-fold decrease in memory usage per core. This project was enabled by a seed investment of around $500,000. Project Leader: Allan D. Grosvenor
• ALYA RED – Barcelona Supercomputing Center developed a first of its kind, in-house, end-to-end biomechanical model including numerical methods, parallel implementation, mesh generation, and visualization. The Alya System is a computational mechanics code with two main features. First, it is specially designed for running with high efficiency in large-scale supercomputing facilities. Secondly, it is capable of solving different physics tasks in a coupled way, each one with its own modeling characteristics: fluids, solids, electrical activity, species concentration, free surface, etc. The Alya Red biomechanical model can help bring drugs to market faster through HPC simulation driven testing. This could result in tens of millions of dollars in potential savings. The long term vision of Alya Red Project is to create an IT infrastructure of hardware and software that can help medical doctors, clinical researchers, and the pharmacological industry to use HPC to positively impact healthcare. The Alya Red simulation relied on Linux driven HPC clusters and IBM BG/P and IBM BG/Q. The interconnect infrastructure included Myrinet and custom interconnect technology. Other tools used included ParaView, Visit, Blender, and Octane. Total investment in developing the infrastructure was around $500,000. Project Leader: Mariano Vazquez

"The Council on Competitiveness would like to congratulate all the winners of the HPC Innovation Excellence Award and thank all of those who submitted entries. The significance of HPC to the private sector will only be fully appreciated when examples such as these are recognized for their economic value," said Dr. Cynthia McIntyre, senior vice president for the HPC Initiative at the Council on Competitiveness.

IDC welcomes award entries from anywhere in the world. Entries may be submitted at any time by completing the brief form available at https://www.hpcuserforum.com/innovationaward/. New winners will be announced multiple times each year. Submissions must contain a clear description of the dollar value or scientific value received in order to qualify. The HPC User Forum Steering Committee performs an initial ranking of the submissions, after which domain and vertical experts are called on, as needed, to evaluate the submissions.

HPC Innovation Excellence Award sponsors include Adaptive Computing, Altair, AMD, Ansys, Appro, Avetec/DICE, the Boeing Company, the Council on Competitiveness, Department of Defense, Department of Energy, Ford Motor Company, Hewlett Packard, HPCwire, Intel, Microsoft, National Science Foundation, NCSA, Platform Computing, Scientific Computing, and SGI.

The next round of HPC Innovation Excellence Award winners will be announced at ISC’13 in June 2013.