A Collaborative Supercomputing Architecture for Geospatial Applications
Dr. Steve Chen
Chairman and CEO of Tonbu Inc.
(Former Sr. Vice President and Chief Architect of Cray Computer)
Abstract
A lot of advances during the last ten years while I was exploring the frontiers of large scale commercial computing, blade computing and collaborative computing as building blocks for future information utility grid environment. Now, I can see the opportunity to not only re-vitalize but also speed up the momentum of scientific, engineering and commercial computing with innovative system architecture, while leveraging more cost-effective and commercially-sustainable hardware and software components. As the volume of geospatial data increases by several orders of magnitude over the next decade, so will the potential for corresponding advances in knowledge of our natural world and in our ability to react to the changes taking place. To meet this grand challenge, we propose an innovative supercomputing architecture which is cost-effective, balanced, scalable, and sustainable. Applying high-performance computing and networking capabilities to human-computer interaction and visualization, we propose to make distributed collaboration as much like same-place work as possible for data analysis and decision making among interdisciplinary groups. This new system architecture will enable the accurate emulation of physical phenomena or processes while allowing intensive human interaction and interpretation. This not only helps traditional PC batch-oriented, compute-intensive applications, but also enables real-time or near real-time collaborative computing among scientists, engineers, and analysts. For example, collaborative medical imaging for early cancer detection and diagnostic would allow radiologists, oncologists, and surgeons to more effectively apply their diverse expertise to a patient to whom they have no direct access. My talk will focus on balanced architecture design, sustainable technology components, and deliverable application performance required to meet the future challenge. Hope my thoughts can stimulate more exciting and innovative approaches for the future supercomputing technology.
Bio
Dr. Chen is recognized throughout the high-performance computing industry as a leader, visionary, entrepreneur, and premier system architect. Dr. Chen currently is the co-founder and Chairman/CEO of Tonbu Inc, a leading collaborative solution and service company for high-tech, electronics, health care, education, entertainment and defense industries. Prior to founding Tonbu, he served as President/CEO of Chen Technology Corporation, focusing on advising top management of large international and small startup companies in assessing their core competencies, market competitiveness, industry trends, customer requirements, product directions, technology roadmaps, and alliance strategies. Before that, Dr. Chen was the EVP/CTO of Product Development and Board Director of Sequent Computer Systems, Inc., a leading provider of large enterprise servers for commercial applications, later acquired by IBM Corp. Dr. Chen initiated and guided Corporate Architecture Council that brought together many CIO’s/Chief Architects of Fortune 2000 companies to exchange and establish Sequent corporate product development strategy. During the 1980’s at Cray Research, Inc., Dr. Chen led his team in developing the most commercially successful parallel vector supercomputers, the Cray X-MP and Y-MP, which accounted for more than 90 percent of Cray Research’s worldwide supercomputer installations and for more than $1 billion in revenue. Within his past 30 years in the computing industry, Dr. Chen has formed many international and multi-corporation strategic partnerships and joint ventures in product and technology development. He has also raised more than US$250M in corporate and private investment capital to found many startup companies that developed and delivered cutting-edge products including high performance supercomputing servers, scalable and reliable commercial servers, computing-on-demand blade servers, and internet-based enterprise collaborative software. Dr. Chen graduated from National Taiwan University with a B.S. in Electrical Engineering and the University of Illinois with a Ph.D. in Computer Science. He currently serves on U.S. National Academy of Engineering, American Academy of Arts and Sciences, and the Boards/Advisers of many high-tech companies in U.S. Silicon Valley. He has published numerous research articles and U.S. patents in high-performance system designs, packaging, parallel architectures, compiler software, and application algorithms development.