Scalable Parallel Libraries Conference October 12-14, 1994 National Science Foundation Engineering Research Center for Computational Field Simulation, Mississippi State, Mississippi Program Announcement and Registration Information Keynote Speaker: ---------------- Mani Chandy - California Institute of Technology, Computer Science & NSF Center for Research in Parallel Computing Invited Speakers: ----------------- Dennis Gannon - University of Indiana Ewing Lusk - Argonne National Laboratory Anthony Skjellum - Mississippi State University David Walker - Oak Ridge National Laboratory Contributed Papers: ------------------- Purushotham V. Bangalore, Anthony Skjellum, and Nathan E. Doss - Scalable Libraries Using MPI: Abstraction, Performance, and Portability Robert Bennett, Kelvin Bryant, Alan Sussman, Raja Das, and Joel Saltz - Jovian: A Framework for Optimizing Parallel I/O Joao Carreira, Luis Silva, and Joao Gabriel Silva - On the Design of Eilean Simon Chapple and Lyndon Clarke - PUL: The Parallel Utilities Library Naindong Fang - PEMPI William Gropp, Edward Karrels, and Ewing Lusk - MPE Graphics -- Scalable X11 Graphics in MPI William Gropp, Lois Curfman McInnes, and Barry F. Smith - Scalable Libraries for Solving Systems of Nonlinear Equations and Unconstrained Minimization Problems Jonathan Hardwick - Porting a Vector Library: a Comparison of MPI, Paris, CMMD and PVM Leslie Hart, Tom Henderson, and Bernardo Rodriguez - A Scalable Parallel Library for Regular Grid-Based Applications Edgar T. Kalns and Lionel M. Ni - Performance Evaluation of Some MPI Implementations on Workstation Clusters Pedro D. Medeiros and Jose' C. Cunha - Is enhancing the functionality of process grouping abstractions in parallel programming libraries desirable and feasible? Natawut Nupairoj and Lionel M. Ni - DaRel: A Portable Data Redistribution Library for Distributed-Memory Machine Mike Osborne - Developments in Mathematical Software for Parallel Computers Ravi Ponnusamy, Rajesh Bordawekar, Rajeev Thakur, Tarvinder Singh, and Alok Choudhary - Passion Runtime Library for Parallel I/O Jeffrey M. Squyres and Andrew Lumsdaine - Waveform Iterative Methods for Parallel Solution of Initial Value Problems John E. West, Michael M. Stephens, and Louis H. Turcotte - Adaptation of Volume Visualization Techniques to MIMD Architectures Using MPI Bwolen Yang and David R. O'Hallaron - Procedure Calls Models for Distributed Parameters in Data Parallel Languages Xinglai Zhuang and Jiangping Zhu - Parallelizing a Reservoir Simulator Using MPI ------------------------------------------------------------------------- Conference fee: $225 for all attendees. Attendance will be limited due to space requirements (approximately 100). Fee includes cost of proceedings (to be distributed after the conference), and some meals. Shuttle service also provided to/from airport during limited times. Fees payable by check to Mississippi State University. PREREGISTRATION IS REQUIRED. ON-SITE REGISTRATION CANNOT BE GUARANTEED. REGISTRANTS ARE STRONGLY ENCOURAGED TO STAY FOR THE ENTIRE THREE-DAY CONFERENCE, TO IMPROVE THE EXCHANGE OF INFORMATION. For further information, contact: Anthony Skjellum, tony@cs.msstate.edu, phone: 601-325-8435 fax: 601-325-8997 Organizing Committee: Anthony Skjellum, Mississippi State University, NSF Engineering Research Center for Computational Field Simulation + Computer Science Department, Donna S. Reese, Mississippi State University, NSF Engineering Research Center for Computational Field Simulation + Computer Science Department, Andrew Lumsdaine, University of Notre Dame, Dept. of Computer Science and Engineering Ewing Lusk, Argonne National Laboratory ---------------------------------------------------------------------------- Hotel Accomodations: Hotel accomodations are available at several hotels close to the NSF Engineering Research Center. The following are reasonable choices: The State House Hotel, (601) 323-2000 The Best Western Motel, (601) 324-5555 The Holiday Inn, (601) 323-6161 Airport: Starkville is served by the nearby Columbus, Mississippi Airport, (GTR), which provides connections through three larger cities: (American/American Eagle) through Nashville, Tennessee, (Delta/ASA) through Atlanta, Georgia, (Northwest Airlink) through Memphis, Tennessee GTR is approximately a 20-minute drive from Starkville, and several rental agencies are on-site. Correct directions (better than rental agencies) will be provided to registered attendees. Shuttle to hotels on October 11/12, and to airport on October 14. Also, shuttle to/from hotels will be provided each morning, and Wednesday, Thursday afternoons. ---------------------------------------------------------------------------- *** Message-Passing Interface Tutorial *** preceding the Scalable Parallel Libraries Conference II (SPLC94) October 11, 1994 National Science Foundation Engineering Research Center for Computational Field Simulation at the <
> Mississippi State, Mississippi The tutorial will be housed at the ERC's new <
>, a Silicon Graphics, Inc. National Training Center, using a state-of-the-art video classroom based on SGI workstations, and SGI Onyx multiprocessor. The classroom systems will provide state-of-the-art interaction between instructors and the attendees, and will also provide the basis for running parallel MPI jobs. Tutorial Instructors: Anthony Skjellum, Ph.D., Nathan Doss, Paula Vaughan, Purushotham Bangalore, Mississippi State University, NSF Engineering Research Center for Computational Field Simulation ---------------------------------------------------------------------------- MPI is the message-passing interface standard, defined by the MPI Forum. This new standard for message-passing incorporates the best aspects of common message-passing practice together with advanced concepts to improve software engineering of large-scale message-passing programs, and to allow access to high-performance hardware of current and soon-to-be-released multicomputers (MPPs) and networked clusters. MPI builds on the experience of systems like P4, PVM, Zipcode, Chameleon, NX, and other systems. MPI builds on the input of vendors: Cray, Intel, IBM, Meiko, Convex, NEC, Kuck and Associates, Parasoft, and others. Intended audience: This tutorial is designed for people who are interested in MPI and applications development, and who plan to attend SPLC94, but who may want an in-depth introduction to MPI prior to attending the conference, which will include/emphasize applications of MPI and libraries built on MPI. The tutorial is targeted as an advanced introduction, so that people who are familiar with C and/or FORTRAN programming will be successful at writing basic MPI codes after the tutorial is completed, but with a broader understanding of the concepts. Ideally, it is aimed at people who have experience with a message-passing system like NX or P4 or PVM, but that is not required. Topics will include: writing simple programs in MPI, porting from other message-passing systems to MPI, and using MPI intermediate and advanced features. Furthermore, basic concepts of application-development in the message-passing environment will be included. The tutorial presenters have extensive experience with the development of MPI as well as participation in the MPI Forum. They have early efforts at developing parallel libraries with MPI as well as applications development based on MPI. NOTE! ----- Paid tutorial attendees will receive a copy of "Using MPI: Portable Parallel Programming with the Message-Passing Interface," by William Gropp, Ewing Lusk and Anthony Skjellum, as part of the registration fee (published by MIT Press). If the book should not be available by October 11, the books will be shipped as soon as available (expected publication is in October, 1994). ---------------------------------------------------------------------------- Tutorial fee: $225 for all tutorial attendees, in addition to SPLC94 conference fees. Class size limit is 10 persons. Attendance at the SPLC94 conference following the tutorial is not required, but is strongly recommended. 8 hour class, followed by 4 hour open laboratory. Lunch included. Shuttle service also provided to/from airport during limited times. Fees payable by check to Mississippi State University. PREREGISTRATION IS REQUIRED. ON-SITE REGISTRATION NOT AVAILABLE. REGISTRATION BY AUGUST 15 IS $225. REGISTRATION BY SEPTEMBER 15 IS $325. REGISTRANTS ARE STRONGLY ENCOURAGED TO STAY FOR THE ENTIRE THREE-DAY CONFERENCE FOLLOWING THIS TUTORIAL IN ADDITION (SEE SPLC94 CALL FOR PARTICIPATION). For further information, contact: Anthony Skjellum, tony@cs.msstate.edu, phone: 601-325-8435 fax: 601-325-8997 ---------------------------------------------------------------------------- For further information on SPLC94, please see SPLC94 announcement. Available by ftp from aurora.cs.msstate.edu; directory: pub/SPLC94/SPLC94.2 On Mosaic: http://www.erc.msstate.ed/ss/splc94.html ---------------------------------------------------------------------------- ------------------------- Registration Form -------------------------------- Scalable Parallel Libraries Conference October 12-14, 1994 National Science Foundation Engineering Research Center for Computational Field Simulation, Mississippi State, Mississippi Name: Address: e-mail: phone: FAX: Attending: ____ Conference _______ MPI Tutorial Please pay in US Dollars by check or money order payable to: Mississippi State University/Engineering Research Center (MSU/ERC) Amount: $225/Conference $225/Tutorial(before September 15) $325/Tutorial(after September 15) Please register by September 15. Registration forms and payment should be mailed to: Scalable Parallel Libraries Conference '94 NSF Engineering Research Center attn: Anthony Skjellum PO Box 6176 Mississippi State, MS 39762 Conference price includes conference banquet (Thursday night) and conference proceedings printed by IEEE Press.