From: morrow@acs.ucalgary.ca (Mike Morrow) Subject: CONF: SS'93 HIGH PERFORMANCE COMPUTING: NEW HORIZONS Date: Sun, 4 Apr 1993 21:10:44 GMT Organization: The University of Calgary, Alberta SS'93 HIGH PERFORMANCE COMPUTING: NEW HORIZONS June 6 - 9, 1993 Palliser Hotel, Calgary, Alberta New Horizons SS'93 extends the traditional supercomputing theme to focus on all aspects of high performance computing, including high performance workstations, integrated environments, related software, communications, and approaching new problems economically. The conference organizers have encouraged attendance by high performance users in industry. We anticipate lively discussion: - Where is the technology going? - What will it do for my organization? - What is the experience in my industry? - How can I implement and manage high performance technologies? Location: The site of SS'93 is located in the heart of Calgary. Close by, you will find the Glenbow Museum; the Centre for Performing Arts where concerts and stage shows take place nightly; and the Stephen Avenue Mall offering six blocks of indoor and outdoor shopping, restaurants, and theatres. For those who would like to arrive earlier, or stay an extra few days, Alberta offers Kananaskis Country, Banff National Park, and Lake Louise -- all within 90 minutes of Calgary. Excellent transportation and accommodation available throughout the area will make your visit enjoyable. Special Sessions Sunday, June 6th 1:00 - 5:00 p.m. Introduction to High Performance Computing Tutorial Sponsored by HPC High Performance Computing Centre (Calgary) The SS'93 introductory tutorial will benefit organizations unfamiliar with supercomputing. This session will demonstrate that high performance computing can enhance organizational competitiveness. A range of current and future technology will be introduced. Monday, June 7th The Emerging Workstation Cluster This tutorial session will introduce participants to the latest concepts in workstation clustering. Technologies and implementation will be discussed. This session will be excellent preparation for the papers and vendor forum presentations on workstation clustering which will directly follow the tutorial. Tuesday, June 8th Operations and Facility Management A High Performance Computing Centre is not just a bigger version of a traditional computer centre. HPC centres' managers face many unique operational issues including managing user expectations, choosing technology, user training, software and application support, and funding. Three speakers will describe their philosophies and experiences. Tuesday, June 8th Forum on Users Needs This 90-minute session spotlights high performance computing user requirements in the next computer generation and beyond. Users from industry, military research, and academic environments will participate in a panel and discussion with vendor representatives, followed by discussion from the floor. Users, managers, software and hardware vendors are encouraged to attend and participate. (May be videotaped). Tuesday, June 8th Debate: The Great War: Fortran vs. C Old Fortran hands already feel cheated by Fortran 90, which incorporates many C features such as pointers and structures; it "no longer looks, nor spells like old FORTRAN." Yet, diehard C evangelists remain steadfast in their attack: Fortran belongs with the dinosaurs. C++ or Smalltalk are also becoming fashionable. And what about the High Performance Fortran effort? Will the trend toward parallel, rather than vector, supercomputers ultimately spell Fortran's death? Tuesday, June 8th Visualization Support Workshop Data visualization facilities are an essential component of any high performance computing environment. Attendees are encouraged to describe their own visualization environments and to discuss issues related to provision of visualization services to a community of users. Support personnel and individual researchers wanting to incorporate visualization methods into their own data analysis are encouraged to attend. Topics of discussion may include: distributed computing, workstation hardware requirements, administration, software, output requirements and services, training, and consulting. Wednesday, June 9th Debate: Big Iron vs. Workstations Vector supercomputers are the workhorses of today's high performance computational research; but, what will be the technology tomorrow? Massively parallel homogeneous systems, mixed technology systems including vector and parallel, or clustered workstations? A panel of distinguished experts will examine the implications of these technologies, and make some predictions for the future. Tuesday Banquet - Science and Supercomputers: Myths and Reality Dr. Larry Smarr Director of the National Center for Supercomputing Applications at Urbana Illinois Supercomputer speeds are expected to double every seven months for the next five years. Such growth is necessary to tackle projects such as mapping the human genome. This talk will examine the current state of the sciences which use High Performance Computing and will look at the potential for scientific breakthroughs. The banquet presentation will be of interest to the general public. Larry Smarr, an internationally recognized astrophysicist, was a pioneer in the movement to dramatically increase the number of academic and industrial researchers using supercomputers in research, development, and manufacturing. He initiated the 1983 proposal to the US National Science Foundation to create a national supercomputer center, and worked with the US Congress in 1984 to establish the current set of four NSF Supercomputer centers and the NSFnet national network. Featured Speakers Jack Worlton is president of Worlton and Associates, a consulting firm specializing in supercomputing. Mr. Worlton will open this year's conference on Monday morning with a provocative talk entitled Perspective 2000. Mr. Worlton spent 30 years with the Los Alamos National Laboratory and has served on numerous national advisory committees for the US government, including committees on supercomputer evaluation and supercomputer centres for the National Science Foundation, the National Research Council, and the Department of Energy. Ralph Roskies' Monday luncheon presentation is entitled Heterogeneous Computing: an NSF Perspective. Dr. Roskies is a founder and the Co-Scientific Director of the Pittsburgh Supercomputing Center (PSC). He is also Professor of Physics at the University of Pittsburgh. In 1984, Dr. Roskies became the first Director of the Certificate Program in Scientific Computing at the University of Pittsburgh, which integrates advanced computing techniques with application programming. David Jefferson founded the field of optimistic synchronization. He invented the first optimistic algorithm, named Time Warp, which implements a very general synchronization abstraction called Virtual Time. Dr. Jefferson has made fascinating contributions to a new field called Artificial Life with his imaginative "programinals" (programmed animals). His Tuesday plenary talk entitled Beyond Massive Parallelism will describe a new model of computation and an associated parallel programming language. Bill Hutchison will speak at the Tuesday luncheon on CANARIE: A Status Update. Mr. Hutchison is an electrical engineer with senior management experience with Honeywell, Consolidated Computer, and in his own management consulting firm, which merged with Ernst & Young in 1989. Mr. Hutchison is president of CANARIE, a not-for-profit consortium established to contribute to the communication infrastructure of Canada in support of effective research, development, and education. Guy Steele, Jr. will address the topic "Is There A Supercomputer in Your Future?" at the Wednesday morning plenary session. An entertaining and provocative speaker, Dr. Steele is a Senior Scientist with Thinking Machines Corporation. Dr. Steele has a wealth of experience in languages, standards, and data parallel algorithms, as well as with compilers, parallel processing, and constraint languages. VENDOR FORUM PRESENTATIONS Alex Informatique Noel Cheeseman, Vice President & C.E.O. Parallel Systems and Work Station Clusters: Making them all work together in an open environment Convex Computer Corporation Greg Astfalk, Senior Staff Scientist What's Lacking in MPP? Parallelism! Cray Research, Inc. (Canada) Vito Bongiorno, Manager of Technical Analyst Support Cray Directions Digital Equipment Corporation Workstation Farms for High Performance Computing Freedman, Sharp and Associates Dan Freedman, Director Introduction to Load Balancer for Workstation Supercomputing Fujitsu Ron Crossan, Canadian National Manager High Performance Computing, Update on Fujitsu High Performance Computing Technology HNSX Supercomputers, Inc. Hypercomp, Lynn Pammett, Director IBM Canada Barry Kaplan, Manager - Scientific and Technical Computing IBM High Performance Computing Directions Kendall Square Research Dr. Nick Camp The KSR1 in Technical Applications nCube Steve Colley, Director Technical Development and Chairman of the Board Pacific-Sierra Research Bob Brode, Senior Product Manager Into and Out Of High Performance Fortran Sun Microsystems Stan Head, Manager, SPARC Technology Marketing The Sparc Vision Sponsors and Exhibitors ACTC Technologies, Inc. Alex Informatique Calgary Council for Advanced Technology Calgary Unix Users' Group Canadian Information Processing Society Convex Computer Corporation Cray Research, Inc. (Canada) Digital Equipment Corporation Freedman Sharp & Associates Fujitsu HNSX Supercomputers, Inc. Hewlett Packard HPC High Performance Computing Centre Hypercomp IBM Canada Kendall Square Research Myrias Technologies nCube Pacific-Sierra Research Parallelogram Research Consortium Inc. Scotia High End Computing SHL Systemhouse Silicon Graphics Inc. Sun Microsystems The University of Calgary Thinking Machines Western Economic Diversification Canada SS'93 REGISTRATION FORM NAME____________________________________________________________ as you would like it to appear on your badge AFFILIATION_____________________________________________________ ADDRESS_____________________________________________________ CITY_______________________ PROVINCE/STATE _____________ COUNTRY _______________________ POSTAL CODE _________________ PHONE ________________________ FAX_________________________ E-MAIL_____________________________________________________ Conference Registration includes one copy of the Proceedings, admittance to all technical sessions, meals, and special events. EARLY REGISTRATION DRAW: Early registrants will be entered into a draw for a prize of one night's accommodation at the Palliser Hotel, valid any time this year (subject to availability). CONFERENCE FEES (includes GST) EARLY LATE AMOUNT Before After May 6 May 6 Individual $250 $290 _________ Block of 10 Registrants $2,000 N/A _________ Student (excludes banquet) $50 $60 _________ Student (includes banquet) $75 $85 _________ Extra Banquet tickets -- $50 _________ Extra proceedings $75 $100 _________ HPC Tutorial (Sunday) $10 $10 _________ (with Conference Reg.) (without Conference Reg.) $25 $25 _________ TOTAL ENCLOSED* _________ *Form of Payment: 1 ____ Cheque in Canadian currency 2 ____ Bill my: ____Visa ____Mastercard account Card No. _______________________ Expires ________ Signature _______________________________________ Withdrawals and cancellations must be made in writing by May 24, 1993, and will be subject to a cancellation fee of $50.00. After this date, no refunds will be available. GST registration # R108102864 Please check ____ if you are interested in receiving more information about the High Performance Computing Tutorial. Please check ____ if you are interested in tourist information. The June 8th Banquet - Science and Supercomputers: Myths and Reality is open to the general public. SEND THIS FORM (OR PHOTOCOPY) AND PAYMENT TO: Registrar SS'93 High Performance Computing: New Horizons c/o Academic Computing Services The University of Calgary 2500 University Drive, N.W. Calgary, Alberta T2N 1N4 CANADA FAX: (403) 282-9199 E-Mail: SS93info@acs.ucalgary.ca -- Mike Morrow MS 742 T2N 1N4 (403) 220-4466