Plenary Speaker Bios
David Jursik is the Worldwide Vice President of Sales responsible for IBM's high-performance computing solutions. The new Deep Computing organization brings together all of IBM's server, storage and systems management solutions to address the comprehensive needs of the global technical computing marketplace in government, higher education, life sciences, industry, business intelligence and digital media. These solutions encompass 32-bit and 64-bit servers, industry-leading disk and tape storage management, high-performance parallel file system software, and common cluster-systems management tools to integrate the operation of IBM's Intel, AMD and Power-based servers. In addition, the Deep Computing OnDemand solution is a new initiative to host large computational jobs on a just-in-time basis, ushering in true OnDemand capacity for high-performance computing requirements.
Jursik previously had worldwide sales responsibility for IBM's Linux cluster solutions. These solutions include IBM's Intel- and AMD-based servers, systems-management software, and integration services. The government, academic, life sciences, automotive, electronics, petroleum exploration and production, and financial-services sectors are all considered key growth segments for Linux clusters. These high-performance computing solutions have proven to be an effective way of lowering the price performance of running quantitative workloads and improving the research and development processes in each of the referenced industries. In addition, Linux clusters have begun to expand into the new application areas of general-purpose compute farms, Web-serving environments, business intelligence and digital media.
Prior to this Jursik was the vice president responsible for e-business solutions sales in the western third of the United States. These solutions included the full suite of products and services required to implement enterprise resource planning, customer relationship management, and supply-chain and e-procurement applications.
Jursik has more than 25 years of related IT and industry sales experience spanning industry, distribution, financial services and government sectors in both commercial and technical applications sales capacities.
Dr. Reza Rooholamini manages the Operating Systems, Cluster, database, and Custom Solutions teams for the Product Group at Dell Computer Corporation. He has been with Dell since 1994. While at Dell, his product-development responsibilities include Dell PowerEdge SMP servers, all cluster product offerings, and custom solutions such as Oracle, SAP and other enterprise applications. From 1991 to 1993 he was responsible for advanced client-server systems at Memorex Telex where he was responsible for development and introduction of a client-server product line. From 1985 to 1991, he was manager of advanced systems with the NCR Network Product Division. Prior to joining NCR, Rooholamini was Professor of Computer Systems at the University of Wisconsin. He has more than 17 years of experience in the computer and the communications industries. He has also served as a technology consultant to numerous companies and has served on several industry-standard committees, which included the OSI Network Management Forum. His current research interests are clustering, client-server and network architecture for multimedia, and Multi-tiered Intranet and Internet architectures. He has more than two-dozen publications in conferences and journals.
Rooholamini obtained a BS in electrical engineering from the University of Illinois at Urbana-Champaign, and an MS in electrical engineering and an MS in computer science from the University of Wisconsin. He also holds a Ph.D. in computer science/engineering from the University of Minnesota.
Brian Ropers-Huilman is the manager of High Performance Computing operations with the Center for Computation and Technology (CCT) at Louisiana State University (LSU). CCT is a new interdisciplinary research center with research groups in advanced technologies and computational sciences, spanning disciplines from engineering, basic and computer sciences, the arts, and business. The center is actively pursuing advanced campus, state, and international networks, and it is currently upgrading its recently acquired 1024-processor Linux cluster.
Ropers-Huilman is responsible for systems software and hardware integration for the high-performance and research clustering systems at CCT. He has 15 years experience in information technology, ranging from systems administration, IT project development and management, and work in HPC for the past 6 years. His current focus is on implementing multiple clusters across a disperse geographic area networked with extremely high-speed interconnects.
Ropers-Huilman has a BS in electrical computer engineering from the University of Wisconsin - Madison, and is pursuing an MBA in technology management from the University of Phoenix.
Tutorials Speaker Bios
Greg Watson received his B.S. in computer science from the University of Tasmania in 1986 and received his Ph.D. from Monash University in 2000. Greg is currently a technical staff member in the Advanced Computing Laboratory at Los Alamos National Laboratory. His research interests focus on tools for parallel computers, program debugging, distributed computing and operating systems.
Greg has been involved in the computing industry for more than 15 years. He worked as a research programmer for the University of Tasmania for two years following his degree, then he spent three years as an IT consultant at Griffith University in Brisbane, Australia. Greg was later hired as manager, Corporate Computer Systems, where he worked for 5 years. In 1994, Greg was appointed for a period of two years to the position of Manager of the Queensland Parallel Supercomputing Facility where he was responsible for establishing and managing the then-largest parallel computing system in Australia. At the end of this appointment, Greg spent the following 12 months as an IT strategic planner in the Information Technology Services department at Griffith University before undertaking his Ph.D. From 2000-2002 Greg worked as a senior research fellow at Monash University.
Ron Minnich is team leader of the Cluster Research Lab at Los Alamos National Laboratory. His most recent work involves the creation of the LinuxBIOS project, as well as working as part of the team that created the Science Appliance software that runs the new 1024-node Pink cluster, as well as many other clusters around the world. Ron's other work is in the area of high performance cluster networking, including the Memory Integrated Network Interface (MINI) and collaboration with Intel Corporation on other advanced networking technologies. Ron received his PhD in Computer Science at the University of Pennsylvania in 1991.
Erik Hendriks is a member of the cluster research team at the Advanced Computing Laboratory. He works on operating systems for Linux clusters and is the designer of the BProc single system image for Linux. He has previously worked on cluster-related projects for NASA and Cray. Erik has a BS and MS in computer science from Johns Hopkins University.
Matthew received his BS degree in 1999 from the University of Oregon in mathematics and computer science with departmental honors. His undergraduate thesis was entitled "An Interoperation and Linking Architecture for Computational Engines". In 2001, he completed his MS degree also at the University of Oregon in computer science. His masters thesis was entitled "The Design of a General Method of Coupling Scientific Simulations". His research focused on formalizing the structure of coupled parallel scientific simulations in relation to control and data flow modeling. Between 1997 and 2001 he was a member of the "ParaDucks" high-performance computing research group at the University of Oregon. In 1999, Matthew spent 6 months working at the Advanced Computing Laboratory at Los Alamos National Laboratory doing research in large-scale cluster monitoring and parallel application coupling. During this time he served in a leadership position as communications director of the LANL Student Association.
In 2001, Matthew became a full-time technical staff member at the LANL ACL. His research focuses on cluster monitoring systems and programming models. He is project leader for the "Supermon" cluster monitoring system and a member of the Cluster Research Team. He also serves as a member of the Common Component Architecture (CCA) project that aims to develop tools and techniques for applying modern component programming techniques to high-performance scientific application codes. His role in this project is in language interoperability support and componentization techniques for Fortran codes.
Felix Wolf was born in Munich, Germany, in 1972. He studied computer science at Aachen University of Technology (RWTH Aachen). After that, he worked on parallel performance tools at Forschungszentrum Juelich in Germany until he obtained his Ph.D. from RWTH Aachen in 2003. In 2001, Felix Wolf visited the IBM T. J. Watson Research Center in New York for an internship. After completion of his Ph.D., he joined the Innovative Computing Laboratory at the University of Tennessee, where he extends his Ph.D. research with hardware counters and user interfaces being focal point.
David Klepacki is a senior staff member with IBM Research. He has more than 20 years of experience in high-performance computing, and has worked in many areas including high-performance processor design, numerically intensive computations, computational physics, parallel computing, application benchmarking, and cluster computing. Currently, he is the Associate Director of the Deep Computing Institute at the IBM T.J. Watson Research Center in Yorktown Heights, NY.
David received his Ph.D. in Theoretical Nuclear Physics from Purdue University in West Lafayette, Indiana. He also has a master's degree in Electrical and Computer Engineering from Syracuse University in New York. He has a strong background and working knowledge in all aspects of high performance computing including processor architecture, application optimization, parallel algorithms, and all popular parallel paradigms. His current interests include performance programming, scalable parallel algorithms, grid technology, and portable high performance computing tools.
Rick Kufrin is a senior member of the technical staff at the University of Illinois' National Center for Supercomputing Applications. He joined NCSA in 1987 and has worked in a number of areas at the center, including high-performance computing, massively parallel processing, artificial intelligence, software tool design, and training/consulting. Rick has authored or co-authored numerous conference papers, technical articles and book chapters covering topics in high-performance computing and AI and is a frequent speaker on these subjects. He holds a master's degree in computer science from UIUC with a focus on artificial intelligence and machine learning. He is the originator and technical lead for the PerfSuite software project that will be covered in this tutorial.
Philip J. Mucci received his Bachelor's Degree in computer science from The Johns Hopkins University in 1993 and his Master's Degree from the University of Tennessee in 1998 under Dr. Jack Dongarra. During this time, he has worked for numerous HPC-related institutions including Thinking Machines, Lockheed Martin, IBM Power Parallel Division and Los Alamos National Laboratories. For his thesis, he developed a fast and portable communication infrastructure for PVM, the well-known parallel computing framework and precursor to MPI. Under funding from the D.o.D. High Performance Computing and Modernization Program, he worked on application optimization, benchmarking, and the development of performance tools. He is the author of several papers and has delivered numerous popular optimization tutorials given at numerous D.o.D. and D.o.E. sites throughout the country. He is the inventor of and technical lead of PAPI, the Performance Application Programming Interface, now in widespread use in compute centers around the world. He is currently a research consultant for the Innovative Computing Laboratory at the University of Tennessee in Knoxville. He is funded in part by the National Science Foundation, the Department of Energy and the Department of Defense. Currently he is on sabbatical at the Center for Parallel Computers at the Royal Institute of Technology in Stockholm, Sweden.
John Towns is the Division Director for Scientific Computing at NCSA. He is the primary NCSA coordinator of activities of the Partners for Advanced Computational Services organization within the Alliance that provide access to and support of high-performance computing resources connected by high-performance networks. Among other activities, he plays significant roles in the deployment and operation of the Alliance computational Grid. These projects embody the deployment of technologies and services to support the establishment of a national advanced computational infrastructure. Towns is a frequent invited presenter on high-performance computing and enabling science and engineering in Grid contexts. He received M.S. degrees in Physics and Astronomy from the University of Illinois at Urbana-Champaign in 1990 and 1991 respectively and a B.S in Physics from the University of Missouri-Rolla in 1987.
Applications Track Speaker Bios
Amy Apon received her Ph.D. in Computer Science from Vanderbilt University in 1994. She is currently an associate professor at the University of Arkansas, Fayetteville. Her research interests include cluster and grid computing, middleware, and low-latency network protocols.
Onur Celebioglu is a systems engineer in the High Performance Cluster Computing Group at Dell Inc. He has been with Dell since 2000. His areas of focus are high-speed networking and HPC interconnects. Celebioglu has a B.S. in electrical engineering from METU in Turkey and an M.S. in electrical and computer engineering from Carnegie Mellon University.
Sue Goudy is a graduate student in the Computer Science Department at New Mexico Institute of Mining and Technology, located in Socorro, a lovely old settlement in the Rio Grande Valley. She supports her educational pursuit by working at Sandia National Laboratories in Albuquerque. One of her tasks there is to ensure that large parallel computing systems are working properly. She would really like to graduate so that she can devote more time to her avocation — growing, spinning, and weaving wool.
Paul Gray is an associate professor in Computer Science at the University of Northern Iowa. His research interests are in high-performance computing environments, high-performance networks, and high-performance computing education at the undergraduate level. He is a National Computational Sciences Institute (NCSI) Fellow, and instructs summer NCSI workshops for undergraduate faculty relating to pedagogical issues of integrating high-performance computing topics into the undergraduate computer science curriculum.
Francisco Junqueira Muniz, Brazilian, works at CDTN (Nuclear Technology Development Center) in Belo Horizonte, Brazil. He is an electronic engineer with a Ph.D. in computer science from the University of Southampton, UK (1994). Research branches include parallel processing, distributed algorithms and dynamic load-balancing approaches.
Bruce Moxon is the Chief Solutions Architect for Panasas, where he works with customers to craft scalable data management solutions around Panasas' novel networked storage architecture. His twenty years of experience in high-performance computing and very large databases includes work on numerous parallel and distributed computing architectures, including IBM SP2s and Linux and Windows compute farms. Prior to Panasas, Moxon was the Sr. Director of Bioinformatics at Perlegen Sciences, where he architected, designed, and implemented a high-throughput computational pipeline and analytical data warehouse to successfully generate and manage Perlegen's 100+ TB human genome variation (SNP) repository.
Ron Westfall is a technical leader in the CTO Office at Cray Canada (previously OctigaBay Systems). He has more than 20 years of computing, telecommunications, software and product development experience. He was the senior system architect for Abatis Systems, which was acquired by Redback Networks in September 2000. In his role at Redback and at Abatis, Westfall was responsible for product requirements, system-level design, technology choices and future product development strategy, including being the principal or contributing inventor on 5 patents pending. Prior to Abatis, Westfall was with Sentry Telecom Systems and MPR Teltech. Westfall holds a bachelor of applied science degree in electrical engineering from the University of Waterloo.
Systems Track Speaker Bios
Ying-Chin Fang is a senior consultant of the Scalable Systems Group at Dell, Inc. He is specialized in large-scale-computing resource management and high-performance computing. He participates in open-source groups and standard organizations as a Dell representative. Yung-Chin's product development responsibilities include Dell HPC clusters, advanced HPC technology R & D and large-scale cyberinfrastructure management. He has more than 14 years of experience in the computer industry and has also served as a technology consultant to numerous companies. He has published more than a dozen computer books in Taiwan and has published works in journals and for numerous conferences. Yung-Chin obtained his BS in computer science from Tamkang University and an MS in computer science from Utah State University.
Joseph Greenseid has been a systems programmer in the Grid and Security Technologies group at NCSA for two and a half years. Some of his most recent projects have involved working on the TeraGrid. Joe has worked with Linux for about seven years and has been interested in clustering for the past three and a half years. He is also the creator and maintainer of a Linux clustering Web site — Linux Clustering Information Center.
James Laros is a principal member of Technical Staff at Sandia National Laboratories. He is currently involved in research, design, development and implementation of System Software for High Performance Computing Systems.
Martin W. Margo received his BA in mathematics-computer science from the University of California, San Diego in 2003. His main interest is high-performance storage systems and high-performance parallel file systems for large scientific clusters. As an HPC systems engineer at San Diego Supercomputer Center (SDSC), he has been working with multi-teraflops Linux clusters at SDSC. He also has been working with systems engineers, Storage Area Network (SAN) engineers, as well as researchers from IBM, Argonne National Laboratory (ANL), and Cluster File Systems (CFS) to perform parallel file systems integration into state-of-the-art production Linux clusters. In 2003, he and a team from SDSC and IBM research won the bandwidth challenge, receiving the Commercial Tools award in Super Computing (SC) 2003 with the entry "On-Demand File Accesss Over a Wide Area with GPFS." Margo, and collaborators from ANL and SDSC, won the Tools award with their entry "High Performance Grid-Enabled Data Movement with GridFTP."
Chona Guiang graduated from the University of the Philippines in 1992 with a B. S. degree in Chemistry. In 1994, she was admitted to The University of Texas at Austin, where she obtained her Master of Arts degree in 1997 and Ph.D. in 1999 in chemical physics. Chona joined the Texas Advanced Computing Center in August of 2000. Since then, she has engaged in several HPC research projects including performance characterization and optimization of applications, and parallelization of a mesoscale lithography simulation code. Chona has also provided consulting support and assisted with TACC's HPC training courses. Her current professional interests include theoretical chemistry and computational bioscience.
Gregory P. Rodgers....
Karl W. Schulz received his Ph.D. from the University of Texas in 1999 in aerospace engineering. After completing a one-year post-doc at the University of Texas, he joined an engineering firm creating commercial computational fluid dynamics (CFD) software. After nearly three years in the commercial CFD industry, he recently returned to the University of Texas as a research associate at the Texas Advanced Computing Center. His research interests include high-performance computing; flow-structure interaction; climate, weather, and ocean modeling; domain decomposition; and unstructured meshing.
Michael Shuey is a Linux developer and administrator at Purdue University's Rosen Center for Advanced Computation. He has worked with Linux for the past eight years and Linux clusters for the past three. Currently he maintains several clusters, ranging from a few nodes to several hundred, to support the work of researchers at Purdue and various partner institutions.
Matthew Woitaszek is a Ph.D. student at the University of Colorado, Boulder. His research interests include performance of cluster and supercomputer systems, hierarchical storage, and data replication for distributed climate modeling workloads.
Cluster Health Speaker Bios
Chokchai Leangsuksun started his career in March 1995 at AT&T's (which later became Lucent) network system. At Lucent he acquired practical experience in telecommunication system reliability and high-availablitily research and development. He led the Lucent Technology R&D team in both technical and project management areas, improving system reliability in several cluster-computing architecture products. He was responsible for improving the downtime of revenue-generating, network-element-provisioning products from 99.5 percent to 99.99 percent. In February 2002, he became an associate professor at Louisiana Tech University where he has taught Entrepreneurship and Information Technology. He founded and co-chaired a high-availability and performance workshop and released the first HA-Beowulf cluster software. He writes articles featured in major technical publications and gives presentations at highly regarded conferences. He has also collaborated with various research groups and national industrial labs, including Oak Ridge National Lab, NCSA, LAM/MPI, Dell, Intel, and Ericsson. In September 2003, he received an outstanding teaching award from the College of Engineering and Science at Louisiana Tech University.
Jim Prewett is a High Performance Computing Systems Engineer at the Center for High Performance Computing at the University of New Mexico. His primary responsibilities there include maintaining high-performance computing resources and networks as well as leading the security team. His primary interests are in security for high performance computing systems and monitoring those systems.
Data Rates Speaker Bios
Thomas Hauser is an assistant professor at the Department of Mechanical and Aerospace Engineering at Utah State University. He received the degrees Diplom-Ingenieur and Doktor-Ingenieur in mechanical engineering from the University of Technology, Munich, in Germany. Prior to coming to Utah State University, he was a postdoctoral research associate at the Department of Mechanical Engineering and at the Center of Computational Science at the University of Kentucky. His cooperation with the cluster group in Electrical and Computer Engineering lead to an Honorable Mention in the category Price/Performance of the Gordon Bell Prize for its work "High-Cost CFD on a Low-Cost Cluster". His research interests are unsteady fluid dynamics simulations, distributed multiphysics simulation, and developing low-cost PC clusters engineered towards his applications.
Kevin Pedretti received his BS in electrical engineering in 1999 and his MS in computer engineering in 2001 from the University of Iowa. As a graduate researcher, he applied high-performance computing techiques to the field of computational genomics. His important contributions included a very high-performance method for clustering EST genetic sequences and a framework for parallelizing NCBI BLAST on clusters of workstations. In 2001, he joined the Scalable Computing Systems group at Sandia National Laboratories. His current research interests include scalable communication protocols for high-performance interconnects, light-weight operating systems for massively parallel computers, and methods for adapting commodity software to the special needs of the HPC community. He is presently developing system software for the upcoming Red Storm supercomputer.
Vendors Speaker Bios
Rebecca Austen is Director, Deep Computing Marketing in IBM Systems Technology Group and is responsible for worldwide marketing for high-performance computing systems. She was previously the marketing leader for IBM Linux clusters and has twenty years of marketing, strategy, development and management experience on Linux, Unix, and PC platforms, systems management, and distributed computing. Rebecca has a B.A. in mathematics from Vassar College.
Robert Ballance is a principal member of Technical Staff at Sandia National Laboratories. He gained his cluster skills while serving as the associate director for Computing Systems and Systems Research at the Center for High-Performance Computing at the University of New Mexico (HPC@UNM). There, he oversaw all of the high-performance computing systems associated with the Center including the 512-processor Los Lobos cluster and the 128-processor RoadRunner cluster. Los Lobos was among the first three Linux clusters to break the top 100 list of supercomputers. At UNM, he also played a guiding role in the early adoption and evolution of Access Grid remote teleconferencing.
Ballance is a member of the ACM, the IEEE Computer Society, and the Internet Society. He serves on the Board of Directors for both the Linux Clusters Institute (LCI) and the New Mexico Information Technology and Software Association. He was the technical program organizer for the premier ClusterWorld Conference and Expo (June 2003) and was a member of the program committee for the 2003 Workshop on Advanced Computing Environments. Bob teaches regularly for the Linux Clusters Institute Workshop series, where he brings depth of knowledge and strong teaching skills to the classroom.
Ballance received his Ph.D. in Computer Science from the University of California, Berkeley, in 1989, where he studied programming environments, user interface design, and operating systems.
Patrick Geoffray is a member of the software development team at Myricom where he authored various middlewares running on GM/Myrinet such as MPICH-GM and VI-GM. More recently, he led the firmware development effort for the Myrinet Express (MX) interface.
Geoffray was born in Lyon, France, and received his Ph.D. in computer science from the University of Lyon in 2001. His interests lie in high-performance computing and high-performance storage.
Ben Rosen ....
Jay Urbanski is a WW HPC IT Architect at IBM specializing in high-performance computing and Linux clusters He was a pioneer of Linux and Linux clusters at IBM and developed many of the IBM "Best Practices" for Linux clusters. He has architected some of the largest Linux cluster supercomputers in the world. Jay has a BA in economics from UT Austin.