TENTATIVE TECHNICAL PROGRAM AND CALL FOR PARTICIPATION PDCS' 94 7th International Conference on Parallel and Distributed Computing Systems October 5 -- 8, 1994 Sahara Hotel Las Vegas, Nevada Sponsored by ISCA, in cooperation with IEEE National Supercomputing Center For Energy and the Environment Northern Telecom CRAY Research Inc. Keynote Speakers: * Ron Ashany National Science Foundation * Albert J. Delorenzi Northern Telecom * Kai Hwang University of Southern California * Michael Flynn Stanford University * Howard J.Seigel Purdue University * Avi Silberschatz University of Texas, Austin * Jeffrey Tsai Universirty of Illinois,Chicago ******************************************************************************** Wednesday, October 5 (Tutorials and Registration) ******************************************************************************** 9:15 - 5:00 1. Parallel Algorithm Design 2. Task Scheduling in Parallel and Distributed Systems 8:15 - 12:00 1. Multidatabase Systems: Adv. Sol. Global Info Sharing Process 2. Concepts & Tools for Development of Distributed Software 1:15 - 5:30 1. Cooperative Information Systems 2. SCI-Based Local Area Multiprocessors (LAMP) 9:00 - 9:00 Registration ******************************************************************************** Thursday, October 6 ******************************************************************************** 8:45 - 9:00 Opening session: Kia Makki Program Chair 9:00 - 10:00 Plennary Address: "Parallel Processing: Past and Furure" Michael J. Flynn (Stanford University) 10:00-10:15 Break 10:15-12:15pm Four Parallel Sessions (1,2,3,4) ------------------------------------------------ Session 1: Parallel Processing Chair: TBA 1. Performing Abstract Interpretation in Parallel Li-Ling Chen (University of Illinois at Urbana-Champaign) W. L. Harrison (University of Illinois at Urbana-Champaign) 2. Improving Performance of Direct-Mapped Caches For Symbolic Applications In Micro-Parallel Processors Ching-Long Su (University of Southern California) Alvin M. Despain (University of Southern California) 3. Architectural Scaling and Analytical Performance Prediction Mark J. Clement (Oregon State University) Michael J. Quinn (Oregon State University) 4. Partitioning the Recirculating Shuffle-Exchange Network Jae-Dong Lee (Kent State University) Kenneth E. Batcher (Kent State University) Session 2: Parallel Databases: I Chair: TBA 1. Parallelization by the Divide-and-Conquer Method in Database Systems Qi Yang (University of Illinois at Chicago) Clement Yu (University of Illinois at Chicago) Chengwen Liu (Depaul University) 2. A Parallel Execution Model for Updating Temporal Databases Avigdor Gal (Technion-Israel Institute of Technology) Opher Etzion (Technion-Israel Institute of Technology) 3. Performance Analysis of the relational project operation for parallel database systems Erich Schikuta (University of Vienna, Austria) Peter Kirkovits (University of Vienna, Austria) 4. Exploiting Shared-Memory Parallel Computers to Parallelize Main Memory Databases Pietro Manzoni (Politecnico di Milano, Italy) Stefano Crespi Reghizzi (Politecnico di Milano, Italy) Session 3: Applications Chair: 1. Merging Arbitrary Data-Flow Structures Employing Hybrid Scheduling Andreas Mitschele-Thiel (Universtat Erlangen-Nurnberg, IMMD VII, Germany) 2. Ensuring Process Persistence: Specification, Implementation and Performability Measure Navin Budhiraja (IBM T.J. Watson Research Center) Madan Gopal (IBM T.J. Watson Research Center) 3. Implementing Lock-Free Queues John D. Valois (Rensselaer Polytechnic Institute) 4. Security and Privacy for Partial Order Time S. W. Smith (Carnegie Mellon University) J. D. Tygar (Carnegie Mellon University) Session 4: Distributed Computing and Systems Chair: 1. A Simple Graph Algorithm for Message-optimized Distributed Computing Dinesh Kadamuddi (University of Illinois, Chicago) Jeffrey J. P. Tsai (University of Illinois, Chicago) 2. On the Design and the Proof of Correctness of a Distributed Hardware Barrier Synchronization Protocol Mukund M. Buddhikot (University of Texas, Austin) Roy Jenevein (University of Texas, Austin) B. F. Womack (University of Texas, Austin) 3. A Distributed Execution Replay for CHORUS Frederic Ruget (Chorus Systems, France) 4. Distributing reactive systems Pual Caspi (VERIMAG) Alain Girault (MERLIN-GERIN) Daniel Pilaud (VERILOG, France) 12:15 - 1:15 Lunch Break 1:15 - 2:15 Plennary Address:"Processor Scheduling in Virtually Owned Computers Paradigm" Avi Silberschatz (University of Texas, Austin) 2:15 - 2:30 Break 2:30 - 4:00 Four Parallel Sessions (5,6,7,8) ------------------------------------------------ Session 5: Task Scheduling: I Chair: 1. An Experimental Evaluation of Task Scheduling on Reconfigurable Multicomputer Architectures W. Addison Woods (George Mason University) H. Douglas Moser (George Mason University) Ophir Frieder (George Mason University) Paul B. Kantor (Rutgers University) 2. Combining Self-Scheduling and Data-Distribution Schemes for Parallel Computations V. A. Saletore (Oregon State University) J. Liu (Oregon State University) 3. Incorporating Job Scheduling for Processor Allocation on Two- Dimensional Mesh-Connected Systems Yung-Kang Chu (Michigan State University) I-Ling Yen (Michigan State University) Diane T. Rover (Michigan State University) Session 6: Mesh Topology Chair: 1. Election on Square Meshes with Link Failures Byungho Yi (Georgia Institute of Technology) Gary L. Peterson (Spelman College) 2. Multicast Trees to Provide Message Ordering in Mesh Networks Javier Cordova (University of Florida) Yann-Hang Lee (University of Florida) 3. Solving the Shortest-Paths Problem on Bipartite Permutation Graphs Efficiently Lin Chen (Fundamental Research Laboratory) Session 7: Parallel Programming Chair: 1. NonVon: Parallel Programming Language Extensions for Imperative Languages Ted Lewis (Naval Postgraduate School) 2. Logic Based and Imperative Coordination Languages Alexander Forst (University of Technology Vienna, Austria) Eva Kuhn (University of Technology Vienna, Austria) Herbert Pohlai (University of Technology Vienna, Austria) Konrad Schwartz (University of Technology Vienna, Austria) 3. Geometric Techniques for Parallelizing and Scheduling Do-loops Tang Dongxing (New Mexico State University) Gopal Gupta (New Mexico State University) Session 8: Hypercube: I Chair: 1. Optimal Hypercube Algorithms for Triangulating Classes of Polygons and Related Problems Danny Z. Chen (University of Notre Dame) 2. A Restricted Subcube Management Scheme for Hypercube Multicomputers Yilong Chen (Texas A&M University) 3. Identifying Maximal Incomplete Subcubes in Faulty Hypercubes Nian-Feng Tzeng (University of Southwestern Louisiana) Guanghua Lin (University of Southwestern Louisiana) 4:00 - 4:15 Break 4:15 - 6:15 Four Parallel Sessions (9,10,11,12) -------------------------------------------------- Session 9: Interconnection Networks Chair: 1. Performance Prediction of Communication Protocols using Deterministic Analysis Raymond E. Miller (University of Maryland, College Park) Zafar Ullah Choudhry (University of Maryland, College Park) 2. Primitive Communication Tasks in MS Networks Emmanouel A. Varvarigos (University of California, Santa Barbara) 3. How Much Adaptivity is Required for Bursty Traffic? Ludmila Cherkasova (Hewlett-Packard Laboratories) Al Davis (Hewlett-Packard Laboratories) Vadim Kotov (Hewlett-Packard Laboratories) Ian Robinson (Hewlett-Packard Laboratories) Tomas Rokicki (Hewlett-Packard Laboratories) 4. Prioritized Conflict Resolution on Multiple Access Broadcast Channel Using Control Mini-Slots W. Melody Moh (San Jose State University) Yu-Jen Chien (San Jose State University) Teng-Sheng Moh (Silicon Valley Research, Inc.) Charles U. Martel (University of California, Davis) Session 10: Task Scheduling: II Chair: 1. The Effect of Operating System Scheduling on High Performance Message Passing Parallel Systems (Short) Ronald Mraz (IBM T.J. Watson Research Center) 2. Distributed Scheduling Support in the Presence of Autonomy (Short) Steve J. Chapin (Kent State University) 3. Task Relocation for Two-Dimensional Meshes Hee Yong Youn (The University of Texas at Arlington) Seong-Moo Yoo (The University of Texas at Arlington) 4. A Split Graph Based Heuristic for Task Allocation in Distributed Systems Hesham H. Ali (University of Nebraska at Omaha) Hesham El-Rewini (University of Nebraska at Omaha) Yinghua Huang (University of Nebraska at Omaha) Invited Talk TBA Session 11A: Parallel Architectures: Chair: 1. Shared-Resource Multithreaded Processor Architecture and Performance Analysis Tim Killeen (Ohio University) Mehmet Celenk (Ohio University) 2. The Impact of Program Structure on the Performance of Scheduling Policies in Multiprocessor Systems Siu-Lun Au (Carleton University, Canada) S. Dandamudi (Carleton University, Canada) 3. Comparison of Arbitration Policies for Cascaded Crossbar Interconnected Multiprocessors Claude Evequoz (Ecole Polytechnique de Montreal, Canada) Session 11B: Parallel Languages: 4. Efficient Allocation of Program Modules On Multicomputers (Short) H. Barada (Lehigh University) N. Adar (Lehigh University) 5. How Many Times Should a Loop Be Unrolled? (Short) Hesham El-Rewini (University of Nebraska at Omaha) Hesham H. Ali (University of Nebraska at Omaha) Session 12: Parallel and Distributed GIS Chair: 1. Distributed Geographical Databases: some specific problems and solutions R. Laurini (INSA - LISI, France) F. Milleret-Raffort (INSA - LISI, France) 2. Effective Parallel Programming for Spatial Analysis: an Experience using a Network Based Approach G. Brunetti (IMA-CNR, Italy) A. Clematis (IMA-CNR, Italy) B. Falcidieno (IMA-CNR, Italy) A. Sanguineti (IMA-CNR, Italy) M. Spagnuolo (IMA-CNR, Italy) 3. Distributing Computations Among GIS Servers M. A. Gennert (University of California, Riverside) N. I. Hachem (Worcester Polytechnic Institute) N. Serrao (Worcester Polytechnic Institute) A. Bansal (Worcester Polytechnic Institute) 4. Load Balancing for Neural-Network Classification of Remote Sesing Data in an Heterogeneous Network of Workstations J. Favela (CICESE Research Center, Mexico) J. Torres (CICESE Research Center, Mexico) H. Hidalgo (CICESE Research Center, Mexico) R. Granillo (CICESE Research Center, Mexico) 7:30 - 8:30 Reception 8:30 - 9:30 Poster Session (list of the papers at the end of the program) ****************************************************************************** Friday, October 7 ****************************************************************************** 9:00 - 10:00 Plennary Address: "High Performance Heterogeneous Computing" Howard Jay Siegel (Purdue University) 10:00 - 10:15 Break 10:15 - 12:15 Three Parallel Sessions (13,14,l5,16) ----------------------------------------------------- Session 13: Parallel Algorithms: I Chair: TBA 1. On the Performance Prediction of Parallel Algorithms Kumar N. Lalgudi (Yale University) D. Bhattacharya (Yale University) 2. On Parallel Divide-and-Conquer Liane Acker (University of Texas, Austin) Robert Browning (University of Texas, Austin) Daniel P. Miranker (University of Texas, Austin) 3. Performance of Parallel Synchronous Branch-and-Bound for 0-1 Integer Programming on Two Parallel Systems Ruurd M. Wiegers (Delft Univ. of Tech., Netherlands) 4. Parallel Implementation of Sparse Simplex Algorithms (Short) Wei Shu (State University of New York at Buffalo) 5. Some Complexity Results for Rings of Petri Nets (Short) Hsu-Chun Yen (National Taiwan University) Bow-Yaw Wang (National Taiwan University) Ming-Sheng Yang (National Taiwan University) Session 14: Hypercube: II Chair: TBA 1. Efficiently Recognizing Free Subcubes on Hypercube Architectures Keith Humenik (Indiana University at South Bend) Peter Matthews (University of Maryland Baltimore County) A.B. Stephens (University of Maryland Baltimore County) Yelena Yesha (University of Maryland Baltimore County) 2. Task Migration in Hypercubes Using all Disjoint Paths Hsing-Lung Chen (National Taiwan Institute of Technology) Nian-Feng Tzeng (University of Southwestern Louisiana) 3. New Dynamic Partitioning Startegy for Hypercube Computers A. Bellaachia (The University of Qatar, Qatar) A. Youssef (The George Washington University) 4. Parallel and Pipelined Parallel Consecutive Sums on a Hypercube with Application to Ray Casting Jianjian Song (National University of Singapore) Renben Shu (National University of Singapore) Session 15: Parallel Architectures Chair: TBA 1. A New Approach to Network Latency Reduction of Multiprocessors by Data Migration in The Absence of Cache Memories Sibabrata Ray (University of Nebraska, Lincoln) Hong Jiang (University of Nebraska, Lincoln) Qing Yang (University of Rhode Island) 2. Time Cost Analysis of Back-Propagation ANNs over a Transputer Network R. S. Tan (The University of Queensland, Australia) V. L. Narasimhan (The University of Queensland, Australia) 3. Evaluating Cache Performance for Vector Processing Tong Sun (The University of Rhode Island) Qing Yang (The University of Rhode Island) 4. Design Methodology for a Parallel Event-Driven Circuit Simulator (Short) Ahmed H. Kamal (University of Louisville) William L. Bradley (University of Louisville) Adel S. Elmaghraby (University of Louisville) 5. SIMD Median-Based Normalizer (Short) W. Robert Bernecky (Naval Undersea Warfare Center) Jose L. Munoz (Naval Undersea Warfare Center) Session 16: Distributed Databases: Cooperative Environments & Techniques Chair: TBA 1. Middleware Support for Heterogeneous Distributed Information M. P. Papzoglou (Queensland University of Technology, Aust.) A. Delis (Queensland University of Technology, Aust.) B. J. Kramer (Fern Universitat Hagen, Germany) 2. Active Database Rules in Distributed Database Systems: A Dynamic Approach in Solving Structural and Semantic Conflicts in Distributed Database Systems Niki Pissinou (University of SW Louisiana) Kay Vanapipat (University of SW Louisiana) 3. Information Repository for Supporting Distributed Information Management Systems James W. Hong (University of Western Ontario) Michael A. Bauer (University of Western Ontario) 4. Localized Decision-Making and the Value of Information in Decentralized Control Edward A. Billard (University of Aizu) Joseph C. Pasquale (University of California, San Diego) Soon M. Chung (Wright State University) 12:15 - 1:15 Lunch 1:15 - 1:45 Plennary Address: "Digital Libraries: Directions from Databases" Ron Ashany (National Science Foundation) 1:45 - 2:15 Panel: "The Role of PDCS in Digital Libraries" Chair: Ron Ashany 2:15 - 2:30 Break 2:30 - 4:00 Four Parallel Sessions (17,18,19,20) ---------------------------------------------------- Session 17: Interconnection Networks Chair: TBA 1. Effectiveness of Message Strip-Mining for Regular and Irregular Communication Akiyoshi Wakatani (Oregon Graduate Institute) Michael Wolfe (Oregon Graduate Institute) 2. Design and Analysis of a Serial Link Interconnection Network Architecture (Short) Betty L. Hickman (University of Nebraska at Omaha) H. R. Sharif (University of Nebraska-Lincoln) 3. Feasible Flows in a Recursive Class of Interconnection Networks Ehab S. Elmallah (University of Alberta, Canada) 4. Virtual Network Resource Partitioning A Multi-domain Network Management Environment (Short) Nasser Modiri (Network Equipment Technologies, Inc.) Session 18: Parallel Algorithms: II Chair: TBA 1. On The Role Of K-Bits Bitonic Sorting Network In Multicast Routing Majed Z. Al-Hajery (Kent State University) Kenneth E. Batcher (Kent State University) 2. On the Bisection Width of the Transposition Network Konstantinos Kalpakis (Univ. of Maryland Baltimore County) Yaacov Yesha (Univ. of Maryland Baltimore County) 3. On the Synchronization-Computation and Dependence- Architecture Trade-offs of Asynchronous Parallel Iterative Algorithms Longsong Lin (National Yunlin Inst. of Technology, Taiwan) Session 19: Parallel Databases : II Chair: TBA 1. Scheduling and Parallelism for Extended SQL Query Optimization A. Hameurlain (Universite Paul Sabatier, Lab. IRIT, France) F. Morvan (Universite Paul Sabatier, Lab. IRIT, France) 2. ASPECT - Specifying Consistency Requirements for Replicated Data from an Applications Point of View Richard Lenz (University of Erlangen-Nuremberg, Germany) Thomas Kirsche (University of Erlangen-Nuremberg, Germany) Berthold Reinwald (University of Erlangen-Nuremberg, Germany) 3. Parallel Off-line Consistency Constraint Checking (Short) David Gadbois (The University of Texas at Austin) Daniel P. Miranker (The University of Texas at Austin) Session 20: Fault Tolerance and Recovery Chair: TBA 1. TBFT: A Task-Based Fault Tolerance Scheme for Distributed Systems Shalini Yajnik (Princeton University) Santhanam Srinivasan (Princeton University) Niraj K. Jha (Princeton University) 2. Distributed Algorithms for the Reconfiguration of Fault-Tolerant Multicomputers Ran Libeskind-Hadas (Harvey Mudd College) 3. Domino-Effect Free Checkpointing Recovery In Distributed Systems Taesoon Park (Texas A&M University) Junguk L. Kim (Texas A&M University) 4:00 - 4:15 Break 4:15 - 6:15 Four Parallel Sessions (21,22,23,24) ---------------------------------------------------- Session 21: Reliability and Recovery Chair: TBA 1. Improvement of Reliability in Hypercubes Using a Fast Reconfiguration Algorithm D. R. Avresky (Texas A&M University) K. Al-Tawil (Texas A&M University) 2. A Heuristic File Assignment Algorithm to Maximize Reliability in a Distributed Computing System Deng-Jyi Chen (National Chiao Tung University, Taiwan) Ruey-Shun Chen (National Chiao Tung University, Taiwan) W. C. Hol (National Chiao Tung University, Taiwan) David T. K. Chen (FORDHAM University) 3. Distributed Computing with a Spreadsheet (Short) Willard Korfhage (Polytechnic University) 4. A Checkpointing & Recovery Scheme for Multiprocessors With Superscalar Processors Mazin S. Yousif (Louisiana Tech University) Session 22: Distributed Algorithms/Mutual Exclusion Chair: 1. An Efficient Fault-Tolerant Mutual Exclusion Algorithm for Distributed Systems D. Manivannan (The Ohio State University) M. Singhal (The Ohio State University) 2. Efficient Deadlock Detection in Distributed Systems Shigang Chen (Florida International Univeristy) Yi Dent (Florida International Univeristy) Cyril Orji (Florida International Univeristy) Wei Sun (Florida International Univeristy) 3. Fault-Tolerant Distributed Mutual Exclusion (Short) Mitchell L. Neilsen (Oklahoma State University) 4. A Phase-Based Mutual Exclusion Algorithm for Computer Network R. Baldoni (Universita di Roma "La Sapienzal") B. Ciciani (Universita di Roma "La Sapienzal") 5. Coping with Different Retrieval Standards in Next Generation Networks David Flater (UMBC/NIST) Yelena Yesha (UMBC/NIST) Session 23: Wormhole Routing Chair: TBA 1. Efficient Algorithms For Non-blocking Wormhole Routing and Circuit Switching on Linear Array Multiprocessors Hong Jiang (University of Nebraska at Lincoln) Joseph Y-T. Leung (University of Nebraska at Lincoln) Ping Luo (University of Nebraska at Lincoln) 2. Performance of Multicast Wormhole Routing Algorithms in Fault Tolerant 2-D Mesh Multicomputers (Short) Fadi N. Sibai (The University of Akron) Sunil D. Kulkarni (The University of Akron) Invited Talk "Shared Memory Processors- CS6400" Mike Lung, CRAY Research, Inc. Invited Talk "Digital Libraries" Yelena Yesha, National Institute of Standards and Technology Session 24: Networks Chair: TBA 1. Higher Order Head-of-Line Blocking and its Impact on the Performance of Interconnection Networks under Uniform and Hot-Spot Traffic M. Jurczyk (Inst. for Microelectronics Stuttgart,Germany) T. Schwederski (Inst. for Microelectronics Stuttgart,Germany) 2. Scalable Interconnection Networks Based on the Petersen Graph Sabine R. Ohring (University of North Texas) Sajal K. Das (University of North Texas) Dirk H. Hohndel (AIB Software Corp.) 3. Broadcasting Trees in Hamming Cubes Sajal K. Das (University of North Texas) Aisheng Mao (University of North Texas) 4. Optimal Asynchronous Agreement and Leader Election Algorithm for Complete Networks with Byzantine Faulty Links (Short) H. Md. Sayeed (Texas A&M University) M. Abu-Amara (Texas A&M University) H. Abu-Amara (Texas A&M University) 7:00 - 9:00 Dinner Keynote Address: "Data Networking in the 1990s" Albert J. Delorenzi, Northern Telecom ****************************************************************************** Saturday, October 7 ****************************************************************************** 9:00 - 10:00 Plennary Address: "Scalable Parallel Processing Systems and Technologies" Kai Hwang (University of Southern California) 10:00 - 10:15 Break 10:15 - 12:15 Three Parallel Sessions (25,26,27,28) ----------------------------------------------------- Session 25: Routing Chair: TBA 1. Hot-Potato Algorithms for Permutation Routing Ilan Newman (The Haifa University, Israel) Assaf Schuster (Technion, Israel) 2. Improving the Performance of Lee's Maze Routing Algorithm on Parallel Computers Ying-Yu Fang (Michigan State University) I-Ling Yen (Michigan State University) Rumi Dubash (University of Houston) Farokh B. Bastani (University of Houston) 3. Improving Internet Routing Information Protocol Tao Wang (Southwest Texas State University) Wuxu Peng (Southwest Texas State University) 4. Deadlock-Free Adaptive Routing in Dual-channel Hypercube Qiang Li (Santa Clara University) Jyh-Chum Liu (Santa Clara University) Session 26: Neural Networks/VLSI Based Architectures Chair: TBA 1. ANN Processing on SIMD Hypercubes Q. M. Malluhi (University of Southwestern Louisiana) M. A. Bayoumi (University of Southwestern Louisiana) T.R.N. Rao (University of Southwestern Louisiana) 2. Residue Systolic Implementations For Neural Networks C. N. Zhang (University of Regina, Canada) M. Wang (University of Regina, Canada) C. C. Tseng (National Chiao-Tung University, Taiwan) 3. A Scaleable Architecture for a Distributed ATM Switch Daniel F. Seidel (University of Southwestern Louisiana) Paul E. Shipley (University of Southwestern Louisiana) Magdy A. Bayoumi (University of Southwestern Louisiana) Invited Talk "CRAY MPP Architecture and Future" Frank Chism, CRAY Research, Inc. Session 27: Performance Evaluation and Modeling Chair: TBA 1. Performance-Based Quality Measures for Parallel Structures Design Robert Todd (The University of Connecticut) Reda A. Ammar (The University of Connecticut) Howard A. Sholl (The University of Connecticut) 2. Modeling and Design Considerations for Handling Inter-Processor Interrupts Suk-Han Yoon (Electronics & Telecom. Research Inst., Korea) Sang-Man Moh (Electronics & Telecom. Research Inst., Korea) Duck-Jin Kim (Korea University, Korea) 3. Architectural Design of Distributed Performance Monitoring Systems: A Hierarchical Approach Jiannong Cao (University of Adelaide, Australia) Olivier de Vel (James Cook University, Australia) Ling Shi (James Cook University, Australia) 4. A Genetic Algorithm Approach for Performance Based Reliability Enhancement of Distributed Systems Anup Kumar (University of Louisville) Sanjay P. Ahuja (SUNY at Oneonta) Session 28: Heterogeneous and Multidatabase Systems Chair: TBA 1. A Performance Study of Concurrency Control in Federated Database Systems Jiandong Huang (Honeywell Technology Center) San-Yih Hwang (University of Minnesota) Jaideep Srivastava (University of Minnesota) 2. Addressing Shared Access and Communication in Distributed Databases using an object oriented environment Suresh Damodaran-Kamal (University of SW Louisiana) Joan M. Francioni (University of SW Louisiana) Niki Pissinou (University of SW Louisiana) 3. Atomic Accesses to Replicated Files in Distributed Systems Xiaohua Jia (The University of Queensland, Australia) 4. Atomic Commitment in Multidatabase Systems Ho-Dong Yoo (Korea Advanced Inst. of Science & Technology) Myoung Ho Kim (Korea Advanced Inst. of Science & Technology) 12:15 -1:15 Lunch 1:15 - 2:15 Plennary Address:" To Be Announced CRAY Research, Inc. 2:15 - 2:30 Break 2:30 - 4:00 Four Parallel Sessions (29,30,31,32) ---------------------------------------------------- Session 29: Parallel Architectures Chair: TBA 1. Maximum Performance Pipelines With Switchable Reservation Tables Ahmed El-Amawy (Louisiana State University) Yi Change Tseng (Louisiana State University) 2. Fast Inner-Product Computation on Short Buses R. Lin (SUNY at Geneseo) S. Olariu (Old Dominion University) 3. An Optimal Mapping Algorithm For HIN-Based Multiprocessor Systems Tissa L. Samaratunga (Wayne State University) Ramaraghavan Srinivasan (Wayne State University) Vipin Chaudhary (Wayne State University) Syed M. Mahmud (Wayne State University) Session 30: Real-Time Systems and Issues Chair: TBA 1. A Nodal Processing Model for Distributed Real-Time Performance Estimation James J. Moriarty (University of Connecticut) Howard A. Sholl (University of Connecticut) James Chen (University of Connecticut) 2. A Priority Ordering-Based Protocol for Concurrency Control in Distributed Real-Time Database Systems (Short) Jinhwan Kim (Seoul National University, Korea) 3. Formal Description of Distributed Multimedia Systems with LOTOS and Real-Time Temporal Logic (Short) Howard Bowman (University of Kent at Canterbury, U.K.) Lynne Blair (Lancaster University, U.K.) Gordon S. Blair (Lancaster University, U.K.) Amanda G. Chetwynd (Lancaster University, U.K.) Keynote Talk "Software Technologies for Distributed Real-Time Computing" Jeffrey Tsai, University of Illinois at Chicago Session 31: Visualization Tools, Languages and Techniques Chair: TBA 1. A Visualization Tool for Display and Interpretation of SISAL Programs Behrooz Shirazi (The University of Texas at Arlington) Hsing-bung Chen (The University of Texas at Arlington) Jessica S. J. Yeh (The University of Texas at Arlington) 2. Performance Visualisation of Message Passing Programs Using Relational Approach (Short) Sute Lei (Macquarie University, Australia) Kang Zhang (Macquarie University, Australia) 3. Visual Occam with Petri Net Semantics M. Al-Mulhem (King Fahd Unv. of Petroleum & Minerals) S. Alis (King Fahd Unv. of Petroleum & Minerals) 4. A Data Parallel Particle Dynamics Toolkit for the Bioblock Modeling System (Short) Edy Cosillo (Southern Illinois University) Michael S.Wainer (Southern Illinois University) Session 32: Interconnection Networks Chair: TBA 1. Multitriangle: A New Interconnection Network Haifeng Qian (Florida Atlantic University) Jie Wu (Florida Atlantic University) 2. Linear Recursive Networks and Their Applications in Distributed Processing Hsu Wen Jing (Nanyang Technological University, Singapore) Moon Jung Chung (Michigan State University) 3. An Algorithm to Locate Switching Faults in General Synchronous Fiber-Optic Networks Marwan H. Abu-Amara (Texas A&M University) Hosame Abu-Amara (Texas A&M University) 4. A New Generalized Star Graph Network: Com-Star (Short) Bin Cong (South Dakota State University) 4:00 - 4:15 Break 4:15 - 6:15 Four Parallel Sessions (33,34,35,36) ---------------------------------------------------- Session 33: Load Balancing Chair: TBA 1. Analysing the Load Balancing Scheme of a Parallel System in a Multiprocessor Environment - A Modeling Approach (Short) Johann Schumann (Technische Universitat Munchen, Germany) Manfred R. Jobmann (Technische Universitat Munchen, Germany) 2. Load Distribution Using Name Sharing for Locally Distributed Systems (Short) Jonathan R. Fausey (Wright State University) 3. Spectral Analysis of Instability in Decentralized Load Balancing (Short) E. A. Billard (University of Aizu, Japan) Invited Talk TBA Invited Talk "Parallel Systems Architecture and Cluster on Parallel Vector Machines" Steve Perronog, CRAY Research, Inc. Session 34: Memory Systems Chair: TBA 1. Coherency Protocol and Algorithm of The DICE Distributed Shared Memory System (Short) Chi-Jiunn Jou (Santa Clara University) Hasan S. AlKhatib (Santa Clara University) Qiang Li (Santa Clara University) Allen Tiekun Chen (Santa Clara University) 2. Modelling Accesses to Stationary Data in a Shared Memory Multiprocessor (Short) Mats Brorsson (Lund University) Per Stenstrom (Lund University) 3. Bandwidth Analysis of Multistage Interconnection Networks for General Memory Reference Model (Short) Shyan-Ming Yuan (National Chiao Tung University, Taiwan) Her-Kun Chang (National Chiao Tung University, Taiwan) 4. A Survey of Parallel Algorithms for Block Bidiagonal Linear Systems on Distributed Memory Machines (Short) M. Paprzycki (University of Texas of the Permian Basin) P. Amodio (Universita di Bari, Italy) T. Politi (Universita di Bari, Italy) Invited Talk "MPP Application Development" Margaret Cahir, CRAY Research, Inc. Session 35: Distributed Systems Chair: TBA 1. An Algorithm for Mutual Exclusion in Distributed Systems Z. Hu (United States Naval Academy) E. K. Park (United States Naval Academy) 2. Rules in Distributed Systems Kia Makki (University of Nevada, Las Vegas) 3. Distributed BFS S.A.M. Makki (University of Queensland, Australia) Invited Talk TBA Session 36: Parallel Algorithms Chair: TBA 1. New Algorithms for Matrix Operations with Applications to a Reconfigurable Parallel Architecture (Short) Yuh-Dauh Lyuu (NEC Research Institute) Eugen Schenfeld (NEC Research Institute) 2. Transitive Closure and Related Algorithms on a Modified PARBS Architecture (Short) Ralph F. Grove, Jr. (University of Louisville) James H. Graham (University of Louisville) 3. Efficient Parallel Selection and Ranking Algorithms in X + Y on Mesh Tarun Anand (I I T, India) Phalguni Gupta (I I T, India) 4. Scalability of a Parallel Hashing Algorithm on a SIMD Machine (Short) Zahira S. Khan (Bloomsburg University) Eugene Kwatny (Temple University) 5. Efficient Parallel Computing Euclidean Distance Transform on Mesh of Trees (Short) Yu-Hua Lee (National Taiwan Inst. of Technology, Tawain) Shi-Jinn Horng (National Taiwan Inst. of Technology, Tawain) Tzong-Wann Kao (National Taiwan Inst. of Technology, Tawain) Yuung-Jin Chen (National Taiwan Inst. of Technology, Tawain) Thursday Poster Papers 1. An Abstract Model of the Parallel Virtual Machine (PVM) Egon Borger (Universita di Pisa, Germany) Uwe Glasser (Universitat-GH Paderborn, Informatik, Germany) 2. Time-Stamp Generation for the Parallel Execution of Program Control Structures Adam Back (University of Exeter, England) Stephen Turner (University of Exeter, England) 3. Authenticated Communication without Relying on Authorities Nobuhisa Fujinami (Sony Computer Science Laboratory Inc.) 4. An Optimal Parallel Algorithm for the Euclidean TSP on EREW PRAM Shao Dong Chen (Griffith University, Australia) Hong Shen (Griffith University, Australia) Francis Suraweera (Griffith University, Australia) 5. The Design of a SliM Image Processor for a SIMD Parallel Architecture for Image Processing Myung Hoon Sunwoo (Ajou University, Korea) Byung Dug Ahn (Ajou University, Korea) Ji Won Jung (Ajou University, Korea) 6. Entice - A Cell Characterization System Using Parallel Computing Binay J. George (Advanced Design Technology, Motorola Inc.) Sean C. Tyler (Advanced Design Technology, Motorola Inc.) Markus G. Wloka (Advanced Design Technology, Motorola Inc.) 7. Using Formal Specification to Aid Negotiating Agents Sandip Sen (University of Tulsa) 8. An Architecture for Homogenizing Federated Databases Kamalakar Karlapalem (Hong Kong University of Sc.& Tech.) Qing Li (Hong Kong University of Sc. & Tech.) Chung-Dak Shum (Hong Kong University of Sc. & Tech.) 9. Formal Parallel Hardware Description Environments Khaled M. Elleithy (King Fahd Univ.Petr. & Minerals) Muhammad A. Al-Humaigani (King Fahd Univ. Petr.& Minerals) 10. A Parallel Scheduler for a Shared Memory (Tightly-Coupled) Multiprocessor System Gopal Sharma (Amdahl Corp.) Bidyut Gupta (Southern Illinois University) 11. Optimisation Methods for Risc machines "In pursuit of lost time" Pierre Christophe Taillebois (L.R.I.-URA 410 CNRS, France) 12. Deadlock-Free Adaptive Routing in Dual-channel Hypercube Qiang Li (Santa Clara University) J Chum Liu (Santa Clara University) 13. Evaluation of Load Sharing Strategies in Distributed Soft Real-Time Systems Based on the Value Function Alfian Budihardjo (STF Technologies, Inc.) Victor Wallace (University of Kansas) 14. Load Balancing Analysis for Farming Transputer Implementation: a Case Study Ioan Dancea (Universite du Quebec a Hull) A. E. Riedmiller (University of Aizu, Japan) ******************************************************************************** REGISTRATION FORM ******************************************************************************** Pre-registration Fee ( received by August 5, 1994) ALL AUTHORS MUST PRE-REGISTER ISCA / IEEE Members U.S. $300.00 Non-Member U.S. $325.00 Student (no paper) U.S. $ 25.00 On-Site Registration Fee (Received after August 5, 1994) ISCA / IEEE Members U.S. $360.00 Non-Member U.S. $380.00 Student (no paper) U.S. $ 30.00 The conference registration fee includes the proceedings, conference reception on Oct. 6, refreshments during the conference, andd luncheon banquet on Oct. 7, 1994. Registratin for student attendees (student/no paper) is for attending the sessions only. Additional reception tickets may be purchased for $15.00 and luncheon tickets may be purchased for $20.00. Make your check or money order payable to ISCA. Be sure to send in the same packet if at all possible: 1: Registration form 2. Registration fee (check or money order) Credit cards are not accepted. Send completed registration packet to: ISCA 8820 Six Forks Road Raleigh, NC 27615 USA If you have any questions, please feel free to contact mas@ncsu.edu. or pdcs@nye.nscee.edu ALL PRE-REGISTRATION MATERIALS MUST BE RECEIVED BY AUGUST 5, 1994. HOTEL INFORMATION We have arranged a block of rooms at the Sahara Hotel, located on the "Strip", 4 miles from McCarran International Airport, 3 miles from Amtrack Station, 3 miles from Downtown, 3 blocks from Las Vegas Convention Center. These rooms have been blocked to give you a special conference rate of $65.00 U.S. single/double and $75.00 U.S. triple occupany. Las Vegas has an 8% room tax applicable to all sleeping rooms. In order to obtain these special rates, please identify yourself with our group: ISCA / PDCS-94 CUT-OFF DATE for RESERVATIONS: September 5, 1994. Please make your reservations early in order to guarantee a room. Telephone toll free 1-800-634-6666 or (702) 737-2654 for reservations. A first night's deposit or credit card guarantee must be received (30) days prior to arrival to hold your reservations on a definite basis. The conference rates are extended three(3) days prior to and after conference dates, subject to availability. Check-in time is 3:00 p.m. and check-out time is 12:00 noon. TRANSPORTATION FROM AIRPORT: Airport Limo: $3.25 Taxi: $10.00 (approx.) The Sahara offers complimentary parking in two valet parking lots and self parking lots. These lots are adjacent to the lobby, the guest rooms, or the casino entrance. The strip location offers an easy walking access to several of the other major hotels and casinos. The hotel is directly adjacent to the new Wet'N Wild theme park, and approximately 1/4 mile from the Fashion Show Mall, featuring 136 of the most dynamic stores in the country, such as Neiman Marcus, Saks Fifth Avenue and Bullocks of California. Looking forward to meeting you in Las Vegas! TUTORIAL REGISTRATION INFORMATION FOR PDCS'94 CONFERENCE: Sahara Hotel, Las Vegas, Nevada October 5, 1994 ----------------------------------------------------------------------------- TUTORIAL REGISTRATION RECEIVED BY AUG. 5: $200.00 per tutorial TUTORIAL REGISTRATION RECEIVED AFTER AUG. 5: $240.00 per tutorial --------------------------------------------------------------------------- Please complete this form (TYPE or PRINT) and return with your payment. Please make check payable to ISCA (check drawn on U.S. bank or international money order) and send to: ISCA Mary Ann Sullivan 8820 Six Forks Road Raleigh, NC 27615 U.S.A. First Name: ___________________________ Last Name: ____________________________ Title (Dr/Ms/Mr): _____________________ Company: ___________________________________Position: __________________ Address: _______________________________________________________________ City: __________________State: ___________________Zip/Postal Code: _________ Country: ____________________ Email: _____________________ Telephone# _____________________ Fax# ____________________________ I'm interested in attending Tutorial Number(s) (please circle your TUTORIAL(S) preference), and my payment is enclosed. 1 2 3 4 5 6 *************** If you must cancel, please notify Mary Ann Sullivan via Fax or E-mail by September 5. After this date no refunds will be made. *************** ****If you have any further questions regarding the tutorials or the conference ****please contact Ms. Mary Ann Sullivan at ISCA headquarters, using the ****following Email, telephone number, or Fax number. Email: mas@ncsu.edu Tel: (919) 847-3747 Fax: (919) 676-0666 Looking forward to seeing you all at the PDCS'94 conference in October. Many Thanks, - Dr. Kia Makki PDCS'94 Program Chair ------------------------------------------------------------------------------ Tutorial # 1 Parallel Algorithm Design A Full Day Tutorial Prof. H. J. Siegel Parallel Processing Laboratory School of Electrical Engineering Purdue University West Lafayette, IN 47907-1285, USA Office: 317-494-3444 Fax: 317-494-6440 Email: hj@ecn.purdue.edu Course Description: ------------------- Parallel machines with 64 to 64,000 processors are commercially available - the challenge is to transform a given task into a parallel algorithm that executes effectively. A variety of techniques for mapping tasks onto large-scale parallel machines are explained and demonstrated through the use of parallel algorithm case studies. Models of SIMD (synchronous), MIMD (asynchronous), and reconfigurable mixed-mode parallel systems are described and contrasted. This tutorial focuses on the design of data-parallel algorithms that can be executed using any of these three modes of parallelism. Issues addressed include choices for data distribution, effect on execution time of increasing the number of processors used (scalability), influence of network topology, use of SIMD vs. MIMD vs. mixed-mode parallelism, trading off computation time vs. communication time, the impact of partitioning the system for subtask parallelism, exploiting overlapping the operation of the SIMD control unit and its processors, and the difficulty of automatic parallelization of serial algorithms. The tasks used for the case studies include window-based image processing, recursive doubling, parallel prefix, global histogramming, 2-D FFT, and sorting. The tutorial concludes with a discussion of some of the ``alligators'' (problems) that make the design and use of large- scale parallel processing systems difficult. Audience: Professionals and researchers with a computer science or computer engineering background who are interested in the use of large-scale parallel processing systems. Brief Biography: H. J. Siegel is a Professor and Coordinator of the Parallel Processing Laboratory in the School of Electrical Engineering at Purdue University. He received two B.S. degrees from MIT, and the M.A., M.S.E., and Ph.D. degrees from Princeton. He has coauthored over 190 technical papers, has coedited six volumes, and has written the book Interconnection Networks for Large-Scale Parallel Processing. He is a Fellow of the IEEE, was a Coeditor-in-Chief of the Journal of Parallel and Distributed Computing, and is currently on the Editorial Boards of both the IEEE Transactions on Parallel and Distributed Systems and the IEEE Transactions on Computers. He was an IEEE Computer Society Distinguished Visitor, was an invited lecturer at a NATO Advanced Study Institute, and is currently a speaker for the ACM Lectureship Series. ----------------------------------------------------------------------------- Tutorial # 2 MULTIDATABASE SYSTEMS: An Advanced Solution for Global Information Sharing Process A Half Day Tutorial (Morning) A.R. Hurson Computer Science and Engineering Department The Pennsylvania State University University Park, PA 16802 E-mail: A2H@ccl.psu.edu ABSTRACT: --------- Users cannot be expected to remember multiple different access methods and access paradigms in order to use separate and existing databases. Nor can users expect all these databases to be converted to a single common model with a single access method. Multidatabases provide users with a common interface to multiple databases with minimal impact on the existing functions of these databases. Multidatabases typically integrate the data from preexisting, heterogeneous local databases into a distributed environment and present global users with transparent methods to use the total information in the system. A key feature is that the individual databases retain their autonomy to serve their existing customer set. Multidatabases, by maintaining the autonomy of local databases, preserve existing organizational investments in local applications and user training, while providing a significant new function of global data accessibility. An important consequence of the local autonomy and heterogeneity of multidatabase systems is that semantically similar pieces of information may have very different names and different data structures in individual local databases. Local-data-access terms are developed to meet specific local requirements and are not globally consistent. Multidatabase designers have created methods to integrate semantically similar, but syntactically different data entities. However, these methods all assume that database designers or users can identify semantically similar entities despite the differences in representation and naming. Without intimate knowledge of the structure of all local databases, this assumption is invalid. It would be interesting if one could develop a scheme that allows automatic identification of semantically similar entities with different access terms. This tutorial explores issues associated with multidatabases and presents the current status of work in this field. It covers a wide range of theoretical, conceptual, and practical topics and addresses the needs of a wide range of audiences, including researchers, database designers, practitioners, and users of databases systems. Audience: The intended audience for this tutorial consists of: (i) Industry professionals and database designers, (ii) Academic professors, and (iii) Students who are involved in research and development in the area of distributed databases. A. R. Hurson is a Computer Engineering Faculty at The Pennsylvania State University. His research for the past 12 years has been directed toward the design and analysis of general as well as special purpose computer architectures. He has published over 130 technical papers in areas including computer architectures, parallel processing, database systems and database machines, dataflow architectures, and VLSI algorithms. Dr. Hurson served as the Guest Co-Editor of special issues of the IEEE Proceedings on Supercomputing Technology, the Journal of Parallel and Distributed Computing on Load Balancing and Scheduling, and the journal of integrated computer-aided engineering on multidatabase and interoperable systems. He is the co-author of the IEEE Tutorials on Parallel Architectures for Database Systems, Multidatabase systems: An advanced solution for global information sharing, Parallel architectures for data/knowledge base systems, and Scheduling and Load Balancing in Parallel and Distributed Systems. He is also the co-founder of the IEEE Symposium on Parallel and Distributed Processing. Professor Hurson has been active in various IEEE/ACM Conferences and has given tutorials for various conferences on database management systems, supercomputer technology, data/knowledge-based systems, scheduling and load balancing, and parallel computing. He served as a member of the IEEE Computer Society Press Editorial Board and an IEEE Distinguished speaker. ----------------------------------------------------------------------------- Tutorial # 3 Task Scheduling in Parallel and Distributed Systems A Full-day Tutorial by Hesham H. Ali and Hesham El-Rewini PDCS'94 Abstract -------- Scheduling is a classical field with several interesting problems and results. A scheduling problem emerges whenever there is a choice as to the order in which a number of tasks can be performed, and the assignment of tasks to servers for processing. A problem may involve jobs that need to be processed in a manufacturing plant, bank customers waiting to be served by tellers, aircraft waiting for landing clearances, or program tasks to be run on a parallel or distributed computer. Clearly, there is a fundamental similarity to scheduling problems regardless of the difference in the nature of the tasks and the environment. In the era of parallel and distributed computing, the scheduling problem has started to re-gain the attention once again. A program can be viewed as a collection of tasks which may run serially or in parallel. The goal of scheduling is to determine an assignment of the tasks to the available processing elements and the order in which the tasks are executed to optimize some performance measures. In this tutorial, we study the task scheduling problem in many of its variations and survey the major related topics including program and system models, optimal algorithms, heuristic algorithms, scheduling non-deterministic programs, task allocation techniques, and software tools. Biography --------- a) Hesham H. Ali Hesham H. Ali is an Associate Professor of Computer Science at the University of Nebraska at Omaha. He has published many articles in the areas of algorithmic graph theory, scheduling, parallel computing and fault-tolerant networks. He has recently co-authored the books "Task Scheduling in Parallel and Distributed Systems," with Hesham El- Rewini and Ted Lewis, 1994; and "Introduction to Graph Algorithms," with Naveed Sherwani, expected 1995. He is also the co-designer of EZG, a new high level language that is developed specifically for implementing graph algorithms, and GRAPHITE, a software package that consists of graph analysis and generation tools. He received his PhD from the University of Nebraska-Lincoln in 1988, and his BS and MS from the University of Alexandria, Egypt, in 1982 and 1985, respectively, all in Computer Science. He is also a member of the ACM. He can be reached at the Department of Computer Science, University of Nebraska at Omaha, Omaha, NE 68182-0500; Internet, hesham@unocss.unomaha.edu b) Hesham El-Rewini Hesham El-Rewini is an assistant professor of computer science at the University of Nebraska at Omaha, and coauthor of two books: 1) Introduction to Parallel Computing (with Ted Lewis; Prentice Hall, 1992) and 2) Task Scheduling in Parallel and Distributed Systems (with Ted Lewis and Hesham Ali; Prentice Hall 1994). He also wrote a chapter on "Partitioning and Scheduling" in the Handbook of Parallel and Distributed Computing (editor: Albert Zomaya, McGraw-Hill, 1994-95). Dr. El-rewini has published dozens of technical papers in the area of parallel & Distributed computing. He contributed to the research literature in 1) parallel programming support environment, and 2) task scheduling. He was a guest coeditor for IEEE Parallel and Distributed Technology's August 1993 issue, and he has been a cochairman of the Software Technology Track of the Hawaii International Conference on System Sciences since 1991. He is also a program committee member of IEEE International Conference on Complex Computer Systems (ICECCS'95). He received his PhD from Oregon State University in 1990, and his BS and MS from the University of Alexandria, Egypt, in 1982 and 1985, respectively, all in computer science. He is a member of the ACM and IEEE Computer Society. He can be reached at the Department of Computer Science, University of Nebraska at Omaha, Omaha, NE 68182-0500; Internet, rewini@unocss.unomaha.edu ----------------------------------------------------------------------------- Tutorial # 4 COOPERATIVE INFORMATION SYSTEMS: MOTIVATIONS, CHALLENGES, POSSIBLE APPROACHES A Half Day Tutorial (Afternoon) Mike P. Papazoglou, Queensland Univ. of Technology, School of Information Systems, Brisbane, Australia Within most organizations, worldwide, mission critical information systems (ISs) already cooperate or are being converted to do so to meet basic business requirements. Due to the lack of appropriate concepts, techniques, and tools, this is being done using primitive and ad-hoc means thereby creating a variety of problems. This talk is devoted to addressing this tidal wave facing the information systems community. The next generation of ISs will involve large numbers of ISs distributed over large, complex computer/communication networks. This ranges from the large and complex billing system of a telephone company, to the small patient information system in a one-doctor office. Such ISs will manage or have access to large amounts of information and computing services. They will support individual or collaborative human work. Information and services will be available in many forms through legacy and new information repositories that support a host of information services. We call such ISs, Cooperative Information Systems (CIS). Soon, the operation of a one-doctor office may critically depend on its ISs' ability to cooperate with foreign ISs not just for reimbursement (i.e., required by insurance organizations) but also for patients (e.g., exchanging information in medical crises). Unlike previous major computing advances based on single technologies, the CIS paradigm will evolve from the integration of many, currently disjoint technologies. Database Systems will contribute information management techniques, particularly for distributed or heterogeneous databases, as well as efficient implementation techniques for information bases. Artificial Intelligence will contribute knowledge representation and reasoning techniques, on the one hand, and distributed problem solving and planning techniques in a multiagent environment on the other. Programming Languages will contribute languages and type/object systems for cooperative programming. Software/Knowledge/Information Engineering will contribute design and development environments/shells and methodologies for CIS development and evolution. Other relevant technologies include: Computer Supported Cooperative Work, Distributed Computing, Organizational Computing, and Interoperability. The challenge is to effectively combine these technologies and their contributions to meet CIS requirements. A significant challenge is to overcome the existing boundaries to achieve a common understanding of the relevant issues. This tutorial will concentrate on the technologies that contribute to the CIS papradigm. Particular emphasis will be given to Distributed Database (DDB) and Distributed Artificial Intelligence (DAI) technologies. These technologies have been conceived to addresses a variety of inherently complex problems involving spatial distribution. Although the types of problems addressed by these two technologies are quite different the technologies themselves present many similarities. The tutorial provides a practical, intuitive, and conceptual understanding of IS cooperation, a vision of the future, current trends / approaches, and research challenges. It addresses such trendy topics as: o Distributed Computing Architectures and Middleware o Distributed Databases o Distributed AI - MultiAgent Systems o Client/Server Environments o Object-Orientation in Databases o Relevant Standards activities such as the ANSWARE, DCE and the Object Management Group (OMG). Short bio ========= Michael Papazoglou as a full Professor and head of School of Information Systems at the Queensland Univ. of Technology (QUT). For two years prior to his appointment at QUT Michael was a Reader at the Australian National Univ. (ANU) in Canberra. Before joining ANU Michael was a senior research scientist and project leader at the German National Research Centre for Computer Science GMD in Bonn, Germany. Michael is the founding editor of the International Journal of Intelligent and Cooperative Information Systems (World Scientific). He is on the editorial advisory board of six international information technology journals and a reviewer for several journals and conferences. His research was funded by several organisations including the Australian Research Council and the European Community. He has consulted to research advisory organisations in USA, Germany, Australia, and the EEC. Michael has authored and edited over 60 books, journal articles, and refereed conference papers. He is a senior member of the IEEE, and a member of the Asia-Pacific Activities Comittee (APAC). He has served as general and program chair in IEEE sponsored conferences, and has been a program committee member of numerous conferences. He has given invited lectures and short courses on Database Technology, Integrating Distributed AI and Database Technologies, Intelligent Information Systems, and Next Generation Database Technology in over a dozen countries. ---------------------------------------------------------------------------- Tutorial # 5 CONCEPTS AND TOOLS FOR DEVELOPMENT OF DISTRIBUTED SOFTWARE A Half Day Tutorial (Morning) Sol M. Shatz Department of Electrical Engineering and Computer Science College of Engineering The University of Illinois at Chicago shatz@eecs.uic.edu Abstract: --------- Concepts and Tools for Development of Distributed Software Distributed computing is widely recognized as a powerful approach for improving performance, reliability, and system growth. While the hardware technology for distributed computing is maturing quickly, the software technology is a serious obstacle to effective use. This tutorial introduces key topics in development of distributed software systems. The major topics covered are specification models and tools; task allocation models and algorithms; distributed programming concepts and languages; and software analysis methods and tools, both for static analysis and dynamic debugging. The participant will learn: 1) specific software development problems that must be considered for effective distributed computing, 2) impacts of design and implementation decisions, 3) programming concepts, and 4) current research trends, methods, and systems. Bio-sketch ---------- Dr. Sol M. Shatz is an Associate Professor in the Department of Electrical Engineering and Computer Science at the University of Illinois at Chicago, where he is the co-director of the Concurrent Software Systems Laboratory. He has been active in teaching and research in the area of distributed computing since 1981 and he has published many journal and conference papers on this topic. Dr. Shatz has prepared two IEEE Computer Society tutorials on Distributed-Software Engineering (one a video-tutorial) and authored the book "Development of Distributed Software: Concepts and Tools" (Macmillan Publishing Co., 1993). He has presented a number of tutorials on topics related to Distributed Software Engineering. Dr. Shatz's research has been supported by AT&T, ONR, and NSF. He received his PhD in 1983 from Northwestern University and is a member of ACM and a Senior member of IEEE. ---------------------------------------------------------------------------- Tutorial # 6 SCI-based Local Area Multiprocessors (LAMP) A Half Day Tutorial (Afternoon) Qiang Li Assistant Professor Department of Computer Engineering Chairman, SCIzzL -- The Association of SCI LAMP Users, Developers and Manufacturers Santa Clara University Santa Clara, CA 95053 (408)554-2730, qli@sunrise.scu.edu ABSTRACT Local Area Multiprocessor (LAMP) is a novel approach to high-performance low-cost parallel processing and high performance networking. A LAMP consists of a large number of personal computers, workstations and other machines connected by a network based on the Scalable Coherent Interface (SCI, IEEE/ANSI Std 1596). A LAMP has the following main characteristics: 1. All nodes (machines) comprising a LAMP share physical memory; 2. Cache coherence between processors is maintained by hardware even when the nodes are distance away; 3. All nodes can access each other's I/O systems directly. The network bandwidth can be as high as 1 Gbyte/sec and the "remote" memory access latency from one node to another can be as low as sub-microseconds. A LAMP can scale up to 64K nodes, and each nodes can contain a number of processors. With today's microprocessor technology, it is feasible to provide 10 Teraops raw computing power. It is important to note that, while high performance LANs can be realized on top of LAMPs, a LAMP is not just another LAN. It is a parallel computer and data transfer is done through shared memory. For example, a simple instruction like "A := B" can now be a network data transfer if the variables A and B are mapped to different nodes. Such a mechanism implies high bandwidth and low latency network data transfer. In this tutorial, we will cover: 1. The basic concept of LAMP; 2. Introduction to SCI; 3. LAMP architecture models; 4. Operating system and application software considerations; 5. The rapid industry acceptance and development of SCI-based systems. ==================================================================== Professor Qiang Li received his B.E. degree in Electrical Engineering from Xi'an Jiaotong University, Xi'an, China, in 1982. He came to the United States in 1982 and earned his M.S. and Ph.D. degrees in Computer Science from Florida International University in 1984 and 1989 respectively. He worked as a System Software Specialist at the Southeast Regional Data Center in Miami, Florida from 1983 to 1989, and became a visiting assistant professor at Florida International University in 1989. Since 1990, he has been an Assistant Professor of the Department of Computer Engineering at Santa Clara University in Santa Clara, California. Dr. Li's research interests include parallel and distributed operating systems, architectures, and programming. His current research is on SCI-based Local Area Multiprocessors (LAMP), especially on the operating system support of SCI LAMP. He co-founded and is chairing the Association of SCI LAMP Users, Developers and Manufacturers which has been the focal point of the SCI technology. Dr. Li is a member of IEEE Computer Society and ACM. He is a member of the IEEE Microprocessor Technical Committee. He has been in the organization committees and program committees of several conferences.