Home
Projects
Software
Bio
 Books
 Papers
 Committees
 Presentations
Students
 Hassan Eslami
 Hormozd Gahvari
 Vivek Kale
 Huong Luu
 Tarun Prabhu
 Xin Zhao

Papers

(This is an incomplete list, containing primarily online papers.)

Papers on MPI and Parallel Programming

Formal methods applied to high-performance computing software design: a case study of MPI one-sided communication-based locking
Salman Pervez, Ganesh Gopalakrishnan, Robert M. Kirby, Rajeev Thakur, and William Gropp, in Software Practice and Experience, 40:1, 2010, pp 23-42.
Load balancing for Regular Meshes on SMPs with MPI
Vivek Kale and William Gropp; at EuroMPI'10
MPI on a Million Processors
Pavan Balaji, Darius Buntinas, David Goodell, William Gropp, Sameer Kumar, Ewing Lusk, Rajeev Thakur and Jesper Larsson Traeff; Outstanding paper at EuroPVMMPI'09, Helsinki, Finland.
Processing MPI Datatypes Outside MPI
Robert Ross, Robert Latham, William Gropp, Ewing Lusk and Rajeev Thakur; Outstanding paper at EuroPVMMPI'09, Helsinki, Finland.
Self-Consistent MPI Performance Guidelines
Jesper Larsson Traeff, William D. Gropp, and Rajeev Thakur, IEEE Transactions on Parallel and Distributed Systems, 21:5, 2009.
Toward Message Passing for a Million Processes: Characterizing MPI on a Massive Scale Blue Gene/P
P. Balaji, A. Chan, R. Thakur, W. Gropp, E. Lusk, ISC Award paper at ISC'09, Hamburg, Germany, 2009.
Non-Data-Communication Overheads in MPI: Analysis on Blue Gene/P
Selected for the Outstanding Papers session at Euro PVMMPI'08, Dublin, Ireland, Sept 7-10, 2008.
Self-consistent MPI-IO performance requirements and expections
William D. Gropp, Dries Kimpe, Robert Ross, Rajeev Thakur, and Jesper Larsson Traeff, in proceedings of EuroPVM/MPI, 2008.
A Portable Method for Finding User Errors in the Usage of MPI Collective Operations
Christopher Falzone, Anthony Chan, Ewing Lusk, and William Gropp. Appeared in International Journal of High Performance Computing Applications, Vol. 21, No. 2, 155-165 (2007)
An Efficient Format for Nearly Constant-Time Access to Arbitrary Time Intervals in Large Trace Files
Anthony Chan, William Gropp, and Ewing Lusk. Appeared in Scientific Programming.
Revealing the performance of MPI RMA Implementations
William D. Gropp and Rajeev Thakur, in proceedings of EuroPVM/MPI, 2007
Extending the MPI-2 Generalized Request Interface
Robert Latham, William Gropp, Robert Ross, and Rajeev Thakur, appeared at EuroPVMMPI 2007.
Nonuniformly Communication Noncontiguous Data: A Case Study with PETSc and MPI
P. Balaji, D. Buntinas, S. Balay, B. Smith, R. Thakur, and W. Gropp, at IPDPS'07.
Advanced Flow-control Mechanisms for the Sockets Direct Protocol over InfiniBand
P. Balaji, S. Bhagvat, D. K. Panda, R. Thakur, and W. Gropp, at ICPP'07.
Open Issues in MPI Implementation
Rajeev Thakur and William Gropp
Test Suite for Evaluating Performance of MPI Implementations That Support MPI_THREAD_MULTIPLE
Rajeev Thakur and William Gropp, at EuroPVMMPI 2007.
A Portable Method for Finding User Errors in the Usage of MPI Collective Operations
Chris Falzone, Anthony Chan, Ewing Lusk, and William Gropp. A "best papers" version of the 2005 EuroPVMMPI paper on collective error checking.
ACM DL Author-ize serviceCollective communication on architectures that support simultaneous communication over multiple links
Ernie Chan, Robert van de Geijn, William Gropp, Rajeev Thakur
PPoPP '06 Proceedings of the eleventh ACM SIGPLAN symposium on Principles and practice of parallel programming, 2006
Issues in Developing a Thread-Safe MPI Implementation
William D. Gropp and Rajeev Thakur, Proceedings of EuroPVMMPI 2006, pages 12-21.
An Interface to Support the Identification of Dynamic MPI 2 Processes for Scalable Parallel Debugging
Christopher Gotbrath, Brian Barrett, William Gropp, Ewing "Rusty" Lusk, and Jeff Squyres
Understanding the Requirements Imposed by Programming Model Middlware on a Common Communication Subsystem
Darius Buntinas and William Gropp. Submitted to Euro PVMMPI'05.
Designing a Common Communication Subsystem
Darius Buntinas and William Gropp. At Euro PVMMPI'05.
Collective Error Detection for MPI Collective Operations
Chris Falzone, Anthony Chan, Ewing Lusk, and William Gropp. At Euro PVMMPI'05.
Improving the Performance of Tensor Matrix Vector Multiplication in Quantum Chemistry Codes
W. D. Gropp, D. K. Kaushik, M. Minkoff, and B. F. Smith.
Implementing MPI-IO Atomic Mode Without File System Support
Robert Ross, Robert Latham, William Gropp, Rajeev Thakur, and Brian Toonen.
Optimizing the Synchronization Operations in MPI One-Sided Communication
Rajeev Thakur, William Gropp, and Brian Toonen.
An Evaluation of Implementation Options for MPI One-Sided Communication,
William Gropp and Rajeev Thakur, at Euro PVMMPI'05.
Predicting Memory-Access Cost Based on Data-Access Patterns
Surendra Byna, Xian-He Sun, William Gropp, and Rajeev Thakur. At IEEE Cluster 2004.
Minimizing Synchronization Overhead in the Implementation of MPI One-Sided Communication
Rajeev Thakur, William Gropp, and Brian Toonen. At EuroPVMMPI'04.
Providing Effcient I/O Redundancy in MPI Environments
William Gropp, Rob Ross, and Neill Miller. At EuroPVMMPI'04
Implementing MPI on the BlueGene/L Supercomputer
George Almasi, Charles Archer,Jose G. Castanos, C. Chris Erway, Philip Heidelberger, Xavier Martorell, Jose E. Moreira, Kurt Pinnow, Joe Ratterman, Nils Smeds, Burkhard Steinmacher-burrow, William Gropp, and Brian Toonen. Appeared as a distinguished paper in the proceedings of Euorpar04.
Design and Implementation of MPICH2 over InfiniBand with RDMA Support
Jiuxing Liu, Weihang Jiang, Pete Wyckoff, Dhabaleswar K Panda, David Ashton, Darius Buntinas, William Gropp, and Brian Toonen. Accepted for IPDPS'04.
High Performance MPI-2 One-Sided Communication over InfiniBand
Weihang Jiang, Jiuxing Liu, Hyun-Wook Jin, Dhabaleswar K. Panda, William Gropp, and Rajeev Thakur. Appeared in the proceedings of the 4th IEEE/ACM International Symposium on Cluster Computing and the Grid (CCGrid 2004), April 2004.
Efficient Structured Data Access in Parallel File Systems
Avery Ching, Alok Choudhary, Wei-keng Liao, Rob Ross, and William Gropp, in the proceedings of Cluster 2003.
Exploring the Relationship Between Parallel Application Run-Time Variability and Network Performance
J Evans, C. Hood, and W. Gropp. In the proceedings of the Workshop on High-Speed Local Networks (HSLN), IEEE Conference on Local Computer Networks (LCN), October, 2003.
MPI on Blue Gene/L: Designing an Efficient General Purpose Messaging Solution for a Large Cellular System
George Almasi, Charles Archer, Jose G. Castanos, Manish Gupta, Xavier Martorell, Jose E. Moreira, William Gropp, Silvius Rus, and Brian Toonen. In the proceedings for Euro PVMMPI'03. Also available as an IBM Research report (PDF).
Implementing Fast and Reusable Datatype Processing
Robert Ross, Neill Miller, and William Gropp. Accepted for Euro PVMMPI'03.
Improving the Performance of Collective Operations in MPICH. (Also available as Postscript.)
Rajeev Thakur and William Gropp. Accepted for Euro PVMMPI'03.
Parallel netCDF: A Scientific High-Performance I/O Interface
Jianwei Li, Wei-keng Liao, Alok Choudhary, Robert Ross, Rajeev Thakur, William Gropp, and Rob Latham. Appeared at SC2003.
Improving the Performance of MPI Derived Datatypes by Optimizing Memory-Access Cost
Surendra Byna, William Gropp, Xian-He Sun, and Rajeev Thakur. Submitted to SC2003.
Noncontiguous I/O Accesses Through MPI-IO
Avery Ching, Alok Choudhary, Kenin Coloma, Wei-keng Liao, Rob Ross, and William Gropp, appeared in the proceedings of CCGrid2003.
Toward Understanding Soft Faults in High Performance Cluster Networks (also available as Postscript).
Jeffrey J. Evans, Seongbok Baik, Cynthia S. Hood, and William Gropp, appeared in the proceedings of the eighth IFIP/IEEE International Symposium on Integrated Network Management.
Fault Tolerance in MPI Programs (also available as Postscript).
William Gropp and Ewing Lusk, in a special issue of the Journal High Performance Computing and Applications
Building Library Components That Can Use Any MPI Implementation (also available as Postscript).
William D. Gropp, appeared at Euro PVMMPI'02
Goals Guiding Design: PVM and MPI (also available as Postscript).
William D. Gropp and Ewing Lusk, appeared at IEEE Cluster 2002.
An Evaluation of a User-Level Data Transfer Mechanism for High-Performance Networks
Philip M. Dickens and William Gropp, appeared at HPDC02.
NIC-Based Atomic Remote Memory Operations in Myrinet/GM
Darius Buntinas, Dhabaleswar K. Panda, and William Gropp, accepted for the Workshop on Novel Uses of System Area Networks (SAN-1).
Learning from the Success of MPI
William Gropp
Scalable Unix Commands for Parallel Processors: A High-Performance Implementation
Emil Ong, Ewing Lusk, and William Gropp
Components and Interfaces of a Process Management System for Parallel Programs
Ralph Butler, William Gropp, and Ewing Lusk.
Runtime Checking of Datatype Signatures in MPI
William Gropp
Exploiting Hierarchy in Parallel Computer Networks to Optimize Collective Operation Performance
N. T. Karonis, B. R. de Supinski, I. Foster, W. Gropp, E. Lusk, and J. Bresnahan.
Reproducible Measurements of MPI Performance Characteristics (Also available as Postscript.)
William Gropp and Ewing Lusk, in the Proceedings of PVMMPI'99.
A Standard Interface for Debugger Access to Message Queue Information in MPI (Also available as Postscript.)
James Cownie and William Gropp, in the Proceedings of PVMMPI'99
Improving the Performance of MPI Derived Datatypes
William Gropp, Ewing Lusk, and Deborah Swider, in the Proceedings of MPIDC'99
ACM DL Author-ize serviceOn implementing MPI-IO portably and with high performance
Rajeev Thakur, William Gropp, Ewing Lusk
IOPADS '99 Proceedings of the sixth workshop on I/O in parallel and distributed systems, 1999
Why are PVM and MPI So Different
William Gropp and Ewing Lusk (The Fourth European PVM - MPI Users' Group Meeting). A longer version entitled PVM and MPI are completely different is also available.
Sowing MPICH: A Case Study in the Dissemination of a Portable Environment for Parallel Scientific Computing
William Gropp and Ewing Lusk (International Journal of Supercomputing Applications, Summer 1997)
An Abstract-Device Interface for Implementing Portable Parallel-I/O Interfaces (also available in Postscript)
Rajeev Thakur, William Gropp, and Ewing Lusk, (Proc. of 6th Symp. on the Frontiers of Massively Parallel Computation, Oct. 1996).
A High-Performance, Portable Implementation of the MPI Message Passing Interface Standard
William Gropp, Ewing Lusk, Nathan Doss, and Anthony Skjellum, (Parallel Computing 1996).
Dynamic Process Management in an MPI Setting
William Gropp and Ewing Lusk. This is the long version of a paper that appeared in the Proc of the 7th IEEE Symposium on Parallel and Distributed Processing, 1995. Also available in Postscript.
Applications-Driven Parallel I/O
N. Galbreath, W. Gropp, and D. Levine, Preprint MCS-P381-0893, September 1993. Appeared in Proceedings of Supercomputing, 1993.
P469.ps.Z and Scalable Unix Tools on Parallel Processors
W. Gropp and E. Lusk, Proc. 1994 Scalable High Performance Computing Conf. (SHPCC '94), IEEE Computer Society Press, 1994, pp. 55-62. Also Preprint MCS-P469-0894.
Scalable Unix Tools on Massively Parallel Processors (also in Postscript).
W. Gropp and E. Lusk. This is a long version of the paper presented at the 1994 Scalable High Performance Computing Conference.
An Introduction to Performance Debugging for Parallel Computers
William Gropp, Preprint MCS-P500-0295, April 1995.
An Experimental Evaluation of the Parallel I/O Systems of the IBM SP and Intel Paragon Using a Production Application
R. Thakur, W. Gropp, and E. Lusk, in Proc. 3rd Int'l Conf. of the Austrian Center for Parallel Computation with special emphasis on Parallel Databases and Parallel I/O, Sept. 1996. Also Preprint ANL/MCS-P569-0296, May 1996.
A Taxonomy of Programming Models for Symmetric Multiprocessors and SMP Clusters(PDF) (also in Postscript)
W. Gropp and E. Lusk
I/O Characterization of a Portable Astrophysics Application on the IBM SP and Intel Paragon
R. Thakur, E. Lusk, and W. Gropp, Preprint MCS-P534-0894, August 1995.
Early Experiences with the IBM SP-1 TM177.ps.Z
edited by William Gropp, Technical Memorandum ANL/MCS-TM-177, June 1993.
Early Experiences with the IBM SP1 and the High-Performance Switch
edited by William Gropp, Technical Report ANL-93/41, November 1993.
A Test Implementation of the MPI Data Message-Passing Standard
William Gropp and Ewing Lusk, Technical Report ANL-92/47, December 1992.

Papers on PETSc, Numerical Libraries, and Domain Decomposition

Hybrid static/dynamic scheduling for already optimized dense matrix factorization
Simplice Donfack, Laura Grigori, William D. Gropp, and Vivek Kale.
Toward Realistic Performance Bounds for Implicit CFD Codes
William Gropp, Dinesh Kaushik, David Keyes, and Barry Smith
ACM DL Author-ize serviceAchieving high sustained performance in an unstructured mesh CFD application
W. K. Anderson, W. D. Gropp, D. K. Kaushik, D. E. Keyes, B. F. Smith
Supercomputing '99 Proceedings of the 1999 ACM/IEEE conference on Supercomputing (CDROM), 1999
Globalized Newton-Krylov-Schwarz Algorithms and Software for parallel implicit CFD
William Gropp, David Keyes, Lois McInnes, and M. D. Tidiri, ANL/MCS-P788-0100
Exploiting Existing Software in Libraries: Successes, Failures, and Reasons Why
William Gropp (presented at the 1998 SIAM meeting on Object Oriented Numerics)
Parallel Newton-Krylov-Schwartz Algorithms for the Transonic Full Potential Equation
Xiao-Chuan Cai, William D. Gropp, David E. Keyes, Robin G. Melvin, and David P. Young
Efficient Management of Parallelism in Object-Oriented Numerical Software Libraries
Satish Balay, Lois Curfman-Mcinnes, William Gropp, and Barry Smith , Also Preprint MCS-P634-0197, 1997.
P356.ps.Z
William D. Gropp and Barry Smith, "The Design of Data-structure-neutral Libraries for the Iterative Solution of Sparse Linear Systems," Preprint MCS-P356-0393.
Why we Couldn't Use Numerical Libraries for PETSc
William Gropp, in the proceedings of the proceedings of the IFIP TC2/WG2.5 Working Conference on the Quality of Numerical Software, Assessment and Enhancement.
Solutions of TEAM Problems 13 and 20 Using a Volume Integral Formulation
Lauri Kettunen, Kimmo Forsman, David Levine, and William Gropp, Preprint MCS-P459-0894, October 1994.
Solutions of TEAM Problem #13 Using Integral Equations in a Sequential and Parallel Computing Environment
Lauri Kettunen, Kimmo Forsman, David Levine, and William Gropp, Preprint MCS-P470-0994, October 1994.
Computational Electromagnetics and Parallel Dense Matrix Computations
K. Forsman, W. Gropp, L. Kettunen, and D. Levine, Preprint MCS-P475-1094, January 1995.
Numerical Simulation of Vortex Dynamics in Type-II Superconductors
W. D. Gropp, H. G. Kaper, G. K. Leaf, D. M. Levine, M. Palumbo, and V. M. Vinokur, J. Comp. Phys., 1996. Also Preprint MCS-P476-1094, November 1994.
Solution of Dense Systems of Linear Euations Arising from Integral Equation Formulations
K. Forsman, W. Gropp, L. Kettunen, D. Levine, and J. Salonen, Preprint MCS-P538-0895, October 1995.
Parallel Computation of Three-Dimensional Nonlinear Magnetostatic Problems
D. Levine, W. Gropp, K. Forsman, and L. Kettunen, Preprint MCS-P558-0196, February 1996.
Volume Integral Equations in Nonlinear 3D Magnetostatics
Lauri Kettunen, Kimmo Forsman, David Levine, and William Gropp, Preprint MCS-P460-0894, September 1994.

Other Papers

ACM DL Author-ize serviceAn adaptive performance modeling tool for GPU architectures
Sara S. Baghsorkhi, Matthieu Delahaye, Sanjay J. Patel, William D. Gropp, Wen-mei W. Hwu
PPoPP '10 Proceedings of the 15th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming, 2010

Talks, Panels, and Tutorials

ACM DL Author-ize servicePerformance modeling as the key to extreme scale computing
William D. Gropp
ICS '11 Proceedings of the international conference on Supercomputing, 2011
ACM DL Author-ize serviceTeaching parallel programming: a roundtable discussion
John Mellor-Crummey, William Gropp, Maurice Herlihy
XRDS: Crossroads, The ACM Magazine for Students - The Changing Face of Programming, 2010
ACM DL Author-ize serviceApplication supercomputing and multiscale simulation techniques
Alice E. Koniges, William Gropp, Ewing Lusk, David C. Eder
SC '06 Proceedings of the 2006 ACM/IEEE conference on Supercomputing, 2006
ACM DL Author-ize serviceAdvanced MPI: I/O and one-sided communication
William Gropp, Ewing Lusk, Rajeev Thakur, Robert Ross
SC '06 Proceedings of the 2006 ACM/IEEE conference on Supercomputing, 2006
Computer Science Department University of Illinois Urbana-Champaign