SGI Technology Powers Computationally Intensive Work in Six Fields of Study Including Biometrics for Security, Computational Physics to Develop New Materials and Human Genome Research -- To establish a Center for High Performance Computing, Florida Atlantic University (FAU) selected high-performance compute and storage systems from Silicon Graphics . FAU, with six campuses in Broward, Palm Beach and St. Lucie counties, will open the Center at FAU Boca Raton. The powerful SGI Altix high performance computer and multi-terabyte SGI InfiniteStorage system were chosen to address the computing requirements of researchers and will enable them to achieve outstanding results and faster time to insight and discovery. The SGI technology will initially be used for six projects: the creation and analysis of advanced tools for parallel program development, biometrics, experimental mathematics, solid-state physics, biomedical science, and media systems development. FAU Jupiter also temporarily houses Scripps Florida, an arm of The Scripps Research Institute in San Diego, which is renowned for biomedical research. While no formal agreement is in place yet, FAU selected the SGI technology for both university research and with an eye toward future collaboration with Scripps Florida. "The proposed Center for High Performance Computing requires a shared-memory architecture to run huge amounts of calculations, run multiple operations, handle the large data sets, including data gathering and dissemination, or work in multi-dimensional space: SGI meets such requirements with the Altix, a shared-memory system with very high processing performance," said Dr. Jie Wu, professor of computer science and engineering, FAU, and chairman, Technical Committee of Distributed Processing, IEEE. "In the long term, this new Center will be very important to South Florida because there is a large concentration of high tech moving into the region -- Motorola, IBM and now Scripps -- and there are many small- and medium-sized technology companies as well. There is a big need for high performance computation." FAU's Charles E. Schmidt College of Science and College of Engineering and Computer Science will initially use the SGI Altix and SGI InfiniteStorage systems and Linux based application software for the following projects: -- Human Genome. To process and interpret the enormous amount of data that has been collected in sequencing the human genome, FAU plans to generate a novel program with a Web-based interface that would allow researchers to identify one or more frequently appearing substrings in a long string. Finding out how these long strings (introns) are spliced out from the messenger RNA will help scientists understand the causes of some genetic diseases and some types of cancer. The project required the SGI Altix and SGI InfiniteStorage systems because of the overwhelming amount of storage and processing needed for multiple operations on a long string. The shared memory of the SGI Altix is also expected to help reduce the complexity of the communication. -- Experimental Mathematics. Researchers in the Department of Mathematical Sciences and the Center for Cryptology and Information Security will use the Altix to investigate algorithmic solutions to several open problems in combinatorics and graph theory. -- Biometrics Computing. Biometrics research, which involves the use of a person's unique biological characteristics such as face, iris, voice, or DNA for personal identification, has applications in personal, corporate and government security. FAU faculty will take advantage of the SGI Altix system's tremendous computing power to perform parallel biometrics computing and develop efficient algorithms that integrate these multiple personal features quickly and efficiently, especially for use with large databases. -- Computational Physics. The FAU faculty will be using the SGI Altix system to simulate the stable and metastable phases of a crystal, stimulating more experimental research for new materials, especially metals. The large, high-bandwidth, low-latency memory capabilities of the SGI Altix system were an absolute requirement and will allow scientists to dramatically increase the size and complexity of materials they simulate. -- Integrated Computation and Communication: A joint project between the Department of Computer Science and Engineering at FAU and EDSS, Inc., a local software company, has resulted in the development of a new, proprietary Technology for Integrated Computation and Communication (TICC). TICC provides facilities for asynchronous message exchange among parallel processes, with message passing latencies as low as 350 nanoseconds. TICC technology was used to develop TICC-PP, a TICC-based Parallel Program development and execution environment. This project will use TICC-PP to develop parallel programs in a variety of application areas using the SGI Altix, which is expected to fully exploit the advantages of TICC-PP in large shared memory multiprocessor environments. TICC-PP advantages include faster execution of parallel programs and ease of parallel program development, maintenance and updating. FAU is seeking to establish TICC-PP as a new standard for high-performance parallel program development and usage. -- Video Coding and Processing. FAU intends to design and develop a series of video coding and processing algorithms, which will run on the SGI Altix. The project, which will generate huge data sets, will focus on three areas: H.264 and MPEG-4 coding techniques; innovative coding techniques, such as an adaptive 3D-DCT; and computer intensive video applications related to HDTV and multimedia security. "FAU's research represents a cross-section of the most critical scientific and engineering research taking place today," said Michael Brown, Sciences segment manager for SGI. "SGI Altix systems and InfiniteStorage solutions give FAU researchers the power they need to tackle these problems and the flexibility they need to pursue new research challenges without replacing their computational infrastructure." Purchased through SGI's exclusive education reseller James River Technical, Inc. (JRTI), FAU installed in July a high-performance SGI Altix 3700 system, with 32 Intel Itanium 2 processors and 32GB memory running Novell's SUSE Linux Enterprise Server, and an SGI InfiniteStorage S300 with 2.8TB storage. The purchase was made possible through support of the National Science Foundation's Major Research Initiative.