APPLICATIONS

By Bill Bell, NCSA -- Companies depend on computing power to design and manufacture new products, reduce costs, and accelerate the speed at which products reach the market. But they also need something more elegant. They need software and applications that properly capture the physics, chemistry, and other scientific principles behind those innovations and improved business practices—codes that simulate complex systems at different scales or that impact different disciplines. Industry currently relies on codes that come from software vendors, academia, and in-house development. Some of them are new, and some are older. And they are rarely integrated, making it difficult to capture the whole picture of what's going on. The high-performance computing community is struggling to address the significant business and technical issues that impede this sort of multidisciplinary and multiphysics modeling. An HPC application software summit was held March 25-26 to explore the creation of a consortium to do just that. The summit was organized by the Council on Competitiveness and the University of Southern California's Information Sciences Institute. It was hosted by the University of Illinois' National Center for Supercomputing Applications (NCSA). More than 100 people representing industries that use high-performance computing, universities, national labs, and hardware and independent software vendors attended. "Simulation-based engineering is fundamental to our nation's leadership—in manufacturing, medicine, security, energy," said Robert Graybill, the Information Science Institute's director of national innovation initiatives. "The community working on the software for simulation-based engineering is small and fragmented. The effort required to make the most of it is large and expensive." "Increasingly companies must conduct multidisciplinary and full life-cycle product simulations to meet the competitive pressures of the global marketplace," explained Suzy Tichenor, vice president of the Council on Competitiveness. "But many companies that use high-performance computing cannot accomplish this task with current application software. This is really an area where the public and private sectors could cooperate to move forward." Lively panel discussions focused on the requirements that end-users in research and development have for multiphysics software and issues surrounding pricing and licensing of that software. There was also extensive discussion of how best to conceive a software framework for integrating codes that must work together to run multiphysics models. Codes written in-house have to work with codes provided by independent software vendors and open-source codes being built by far-flung communities. A software framework could be the solution. "The right solution will raise all manufacturing boats, it will expand markets for software vendors. It will improve the sort of basic research that universities and national labs engage in while helping push that research out into the sectors that apply it. This is a seminal discussion for everyone involved," said Merle Giles, director of NCSA's Private Sector Program. The summit organizers plan to spend the next few months updating the concept paper prepared for the summit, clarifying the technical and business challenges that a consortium would tackle, exploring how the consortium would be organized, and formalizing interest from institutions from all sectors. They plan to have the first organizational meeting of interested members this summer. The HPC Application Software Consortium Summit was sponsored by ATK, the Council on Competitiveness, IBM, MSC Software, NASA, NCSA, the National Science Foundation, Procter & Gamble, and USC's Information Sciences Institute. For further information, contact: Bob Graybill, USC Information Sciences Institute, rgraybill@isi.edu Merle Giles, NCSA Private Sector Program, mgiles@ncsa.uiuc.edu Suzy Tichenor, Council on Competitiveness, stichenor@compete.org