Papers/Publications/Presentations

Please note

The copyrights of the following material are held by the publishers. The downloadable files (if given) are preprints. Please treat this material in a way consistent with the "fair use"-provisions of the appropriate copyright owners.

This page contains the publications of the years 2008, 2007, 2006, 2005 and 2004. Older publications can be found below. For a list dedicated to Computer Architecture Technical Reports see below.

2008

Andreas Heinig and Jochen Strunk and Wolfgang Rehm and Heiko Schick:: Generalizing the SPUFS concept- a case study towards a common accelerator interface.
Accepted for publication in proceedings of the Many-core and Reconfigurable Supercomputing Conference (MRSC) 2008
April 2008, The Queen's University of Belfast, Belfast, Northern Ireland.
Torsten Hoefler and Timo Schneider and Andrew Lumsdaine:: Accurately Measuring Collective Operations at Massive Scale.
Accepted for publication at the 7th International Workshop on Performance Modeling, Evaluation, and Optimization of Ubiquitous Computing and Networked Systems (PMEO-UCNS 2008) at the IPDPS08
Timo Schneider, Torsten Hoefler, Simon Wunderlich, Torsten Mehlan, Wolfgang Rehm:: An optimized ZGEMM implementation for the Cell BE
In Proceedings of the 9th Workshop on Parallel Systems and Algorithms (PASA) 2008, February 26, 2008 in Conjunction with 21st International Conference on Architecture of Computing Systems (ARCS)
Dresden, Germany, February 25th to 28th, 2008, ISSN: 1617-5468, ISBN: 978-3-88579-218-5
(PDF Document)

2007

Torsten Hoefler, Marek Mosch, Torsten Mehlan, Wolfgang Rehm:: CollGM - A Myrinet/GM optimized collective component for Open MPI
In Proceedings of the 3rd Workshop KiCC (Kommunikation in Clustern und Clusterverbundsystemen), 2007, Aachen, Germany, URN: urn:nbn:de:hbz:82-opus-21137
(PDF Document)
Frank Mietke, Torsten Mehlan, Torsten Hoefler, Wolfgang Rehm:: Design and Evaluation of a 2048 Core Cluster System
In Proceedings of the 3rd Workshop KiCC (Kommunikation in Clustern und Clusterverbundsystemen), 2007, Aachen, Germany, URN: urn:nbn:de:hbz:82-opus-21137
(PDF Document)
Torsten Hoefler and Andrew Lumsdaine and Wolfgang Rehm:: Implementation and Performance Analysis of Non-Blocking Collective Operations for MPI
In Proceedings of Supercomputing 2007 (SC07), Reno, USA, November 2007
(PDF Document)
Torsten Hoefler and Prabhanjan Kambadur and Richard L. Graham and Galan Shipman and Andrew Lumsdaine:: A Case for Standard Non-Blocking Collective Operations
In Proceedings of EuroPVM/MPI 2007, Paris, France, October 2007, Springer, ISSN: 0302-9743, ISBN: 978-3-540-75415-2
(PDF Document)
Timo Schneider and Simon Wunderlich and Wolfgang Rehm and Torsten Hoefler and Heiko Schick:: Code optimization for Cell/B.E. - Opportunities for ABINIT - a software package for physicists
Poster at the IBM CAS Software and Systems Engineering Symposium
October, 2007, Dublin, Ireland
(PDF Document)
Andreas Heinig and Jochen Strunk and Wolfgang Rehm and Heiko Schick:: Heterogeneous Multiprocessing - On a tightly coupled Opteron Cell evaluation platform
Poster at the IBM CAS Software and Systems Engineering Symposium
October, 2007, Dublin, Ireland
(PDF Document)
Frank Mietke:: Erfahrungen mit parallelen Dateisystemen
Presented in Chemnitz, Germany, September 2007, Megware HPC Users Meeting
(PDF Document)
Torsten Hoefler and Torsten Mehlan and Andrew Lumsdaine and Wolfgang Rehm:: Netgauge: A Network Performance Measurement Framework
In Proceedings of High Performance Computing Conference (HPCC), Houston, USA, September 2007
(PDF Document)
Frank Mietke:: Erfahrungsberichte - HPC im akademischen Umfeld CHiC - TU Chemnitz
Presented in Böblingen, Germany, June 2007, IBM HPC Clusterworkshop
(PDF Document)
Torsten Hoefler and Peter Gottschling and Andrew Lumsdaine and Wolfgang Rehm:: Optimizing a Conjugate Gradient Solver with Non-Blocking Collective Operations
In the Elsevier Journal of Parallel Computing (PARCO), Vol 33, September 2007, SSN: 0167-8191
(PDF Document)
Frank Mietke, Torsten Hoefler, Torsten Mehlan and Wolfgang Rehm:: Diskless Cluster und Lustre - Erfahrungsbericht zum CHiC
Presented in Chemnitz, Germany, April 2007, UNIX Stammtisch 24.04.2007
(PDF Document)
Frank Mietke, Torsten Mehlan, Torsten Hoefler and Wolfgang Rehm:: Stand HPC Cluster CHiC
Presented in Leipzig, Germany, April 2007, ZKI Arbeitskreis Supercomputing 19.04.2007
(PDF Document)
Frank Mietke:: Diskless Cluster und Lustre - Erfahrungsbericht zum CHiC
Presented in Chemnitz, Germany, March 2007, Chemnitzer Linux-Tage 2007
(PDF Document)
Torsten Hoefler, Andre Lichei and Wolfgang Rehm:: Low-Overhead LogGP Parameter Assessment for Modern Interconnection Networks
In Proceedings of the 21st IEEE International Parallel & Distributed Processing Symposium IPDPS 2007, Long Beach, CA, USA, Mar. 2007, ISBN: 1-4244-0909-8
(PDF Document)
Torsten Hoefler, Christian Siebert and Wolfgang Rehm:: A practically constant-time MPI Broadcast Algorithm for large-scale InfiniBand Clusters with Multicast
In Proceedings of the 21st IEEE International Parallel & Distributed Processing Symposium IPDPS 2007, Long Beach, CA, USA, Mar. 2007, ISBN: 1-4244-0909-8
(PDF Document)
Frank Mietke, Dirk Dunger, Torsten Mehlan, Torsten Hoefler and Wolfgang Rehm:: A native InfiniBand Transporter for MySQL Cluster
In Proceedings of the 2nd Workshop "Kommunikation in Clusterrechnern und Clusterverbundsystemen" (KiCC'07), Chemnitz, Germany, February 8th, 2007 ISSN 0947-5125
(PDF Document)

2006

Torsten Hoefler, Rebecca Janisch and Wolfgang Rehm:: Parallel scaling of Teter's minimization for Ab Initio calculations
In Proceedings, HPCnano Workshop 2006 held in conjunction with the SC06 International Conference for High Performance Computing, Networking, Storage and Analysis, Tampa, USA, November 2006
(PDF Document)
Torsten Hoefler, Jeffrey Squyres, Wolfgang Rehm and Andrew Lumsdaine:: A Case for Non-Blocking Collective Operations
In Proceedings, Frontier on High Performance Computing and Networking (FHPCN-06), associated with the ISPA-06, Sorrento, Italy, December 2006, ISBN: 978-3-540-49860-5
(PDF Document)
Wolfgang Rehm:: Zur HPC-Clustercomputer-Beschaffung CHiC
Vortrag zum 26. Treffen des ZKI Arbeitskreises Supercomputing, LRZ München, 19./20.10.2006.
Torsten Hoefler, Jeffrey Squyres, Graham Fagg, George Bosilca, Wolfgang Rehm and Andrew Lumsdaine:: A New Approach to MPI Collective Communication Implementations
In Proceedings, Distributed and Parallel Systems - From Cluster to Grid Computing, presented in Innsbruck, Austria, pages 45-54, Springer, ISBN: 978-0-387-69857-1, Sep. 2006
(PDF Document)
Torsten Hoefler, Peter Gottschling, Wolfgang Rehm and Andrew Lumsdaine:: Optimizing a Conjugate Gradient Solver with Non Blocking Collective Operations
In Proceedings, EuroPVM/MPI 2006, special session ParSim 2006, Bonn, September 2006, ISSN: 0302-9743, ISBN: 3-540-39110-X
(PDF Document)
Torsten Mehlan, Jochen Strunk, Torsten Hoefler, Frank Mietke and Wolfgang Rehm:: IRS - A portable Interface for Reconfigurable Systems
In Proceedings, 5th International Symposium on Parallel Computing in Electrical Engineering, Bialystok, September 2006, ISBN: 0-7695-2554-7
(PDF Document)
Torsten Hoefler, Carsten Viertel, Torsten Mehlan, Frank Mietke and Wolfgang Rehm:: Assessing Single-Message and Multi-Node Communication Performance of InfiniBand
In Proceedings, 5th International Symposium on Parallel Computing in Electrical Engineering, Bialystok, September 2006, ISBN: 0-7695-2554-7
(PDF Document)
Torsten Hoefler, Mirko Reinhardt, Torsten Mehlan, Frank Mietke and Wolfgang Rehm:: Low Overhead Ethernet Communication for Open MPI on Linux Clusters
CSR-06-06, July, 2006, Chemnitz, ISSN 0947-5125
(PDF Document)
Frank Mietke, Robert Rex, Robert Baumgartl, Torsten Hoefler, Torsten Mehlan and Wolfgang Rehm:: Analysis of the Memory Registration Process in the Mellanox InfiniBand Software Stack
In Proceedings, European Conference on Parallel Computing (Euro-Par 2006), Dresden, September 2006, ISBN: 3-540-37783-2
(PDF Document)
Robert Rex, Frank Mietke, Christoph Raisch, Hoang-Nam Nguyen and Wolfgang Rehm:: Improving Communication Performance on InfiniBand by Using Efficient Data Placement Strategies
In Proceedings, International Conference on Cluster Computing (Cluster 2006), Barcelona, September 2006, ISBN: 1-4244-0328-6
(PDF Document)
Torsten Hoefler, Torsten Mehlan, Frank Mietke and Wolfgang Rehm:: LogfP - A Model for small Messages in InfiniBand
In Proceedings, 20th International Parallel and Distributed Processing Symposium IPDPS 2006 (PMEO-PDS 06), Greece, April 2006, ISBN: 1-4244-0054-6
(PDF Document)
Torsten Hoefler, Torsten Mehlan, Frank Mietke and Wolfgang Rehm:: Fast Barrier Synchronization for InfiniBand
In Proceedings of the Workshop on Communication Architecture for Clusters (CAC06), held in conjunction with
IEEE International Parallel and Distributed Processing Symposium (IPDPS), Rhodes Island, Greece, April 2006, ISBN: 1-4244-0054-6
(PDF Document)
Torsten Hoefler, Torsten Mehlan, Frank Mietke and Wolfgang Rehm:: Adding Low-Cost Hardware Barrier Support to Small Commodity Clusters
In Proceedings of 19th International Conference on Architecture and Computing Systems - ARCS'06, Frankfurt, Germany, March 2006, ISSN: 3-88579-175-7
(PDF Document)
2005
Torsten Hoefler, Rebecca Janisch and Wolfgang Rehm:: A Performance Analysis of ABINIT on a Cluster System
Published in Lecture Notes in Computational Science and Engineering, Springer, ISBN 3-540-33539-0, 200 5
(PDF Document)
Wolfgang Rehm (Ed.):: Kommunikation in Clusterrechnern und Clusterverbundsystemen
Tagungsband zum 1. Workshop Kommunikation in Clusterrechnern und Clusterverbundsystemen (KiCC 2005), CSR-05-03, Chemnitz, Deutschland, November 29th, 2005, ISSN 0947-5125
Torsten Hoefler, Torsten Mehlan, Frank Mietke, Wolfgang Rehm:: A practical Approach to the Rating of Barrier Algorithms using the LogP Model and Open MPI
In Proceedings of the International Workshop on Performance Evaluation of Networks for Parallel, Cluster and Grid Computing Systems (PEN-PCGCS'05) in conjunction with the 2005 International Conference on Parallel Processing (ICPP-05), Univ. of Oslo, Norway, June 2005, ISBN: 0-7659-2381-1
(PDF Document)
Torsten Mehlan, Torsten Hoefler, Frank Mietke and Wolfgang Rehm:: Integration of the SISCI Shared Memory Interface into Open MPI
In Proceedings of the 1st Workshop ''Kommunikation in Clusterrechnern und Clusterverbundsystemen'' (KiCC'05), CSR-05-03, Chemnitz, Germany, November 29th, 2005 ISSN 0947-5125
(PDF Document)
Frank Mietke, Robert Rex, Torsten Hoefler, Torsten Mehlan and Wolfgang Rehm:: Reducing the Impact of Memory Registration in InfiniBand™
In Proceedings of the 1st Workshop ''Kommunikation in Clusterrechnern und Clusterverbundsystemen'' (KiCC'05), CSR-05-03, Chemnitz, Germany, November 29th, 2005 ISSN 0947-5125
(PDF Document)
Torsten Hoefler, Jeffrey M. Squyres, Torsten Mehlan, Frank Mietke and Wolfgang Rehm:: Implementing a Hardware-Based Barrier in Open MPI
In Proceedings of the 1st Workshop ''Kommunikation in Clusterrechnern und Clusterverbundsystemen'' (KiCC'05), CSR-05-03, Chemnitz, Germany, November 29th, 2005 ISSN 0947-5125
(PDF Document)
Torsten Hoefler und Wolfgang Rehm: A Communication Model for Small Messages with InfiniBand™
In Proceedings of the Parallel-Algorithmen, -Rechnerstrukturen und -Systemsoftware (PARS) Workshop 2005, Lübeck, June 2005, ISSN 0177-0454
(PDF Document)
Frank Mietke, Marco Steiger, Torsten Mehlan, Torsten Hofler und Wolfgang Rehm: SHIBA Shared Memory Support for InfiniBand™ MPICH2 Device
In Proceedings of the Parallel-Algorithmen, -Rechnerstrukturen und -Systemsoftware (PARS) Workshop 2005, Lübeck, June 2005, ISSN 0177-0454
(PDF Document)
2004
Torsten Mehlan, Wolfgang Rehm, Ralph Engler, Tobias Wenzel:: Providing a High-Performance VIA-Module for LAM/MPI In proceedings of PARELEC'04 2004 IEEE International Conference on Parallel Computing in Electrical Engineering, September 7-10 2004, Dresden, Germany, ISBN: 0-7695-2080-4
This paper is copyrighted by IEEE!
(PDF Document) (c) 2004 IEEE
Christian Siebert:: One-sided Synchronization - A new Approach to Synchronization Primitives
Computer Architecture Technical Report, Chemnitz University of Technology, Dept. of Computer Science, July 2004
(PDF Document).
Torsten Hoefler:: Meta Analysis of Gigabit Ethernet over Copper Solutions for Cluster-Networking
Computer Architecture Technical Report, Fakultaet fuer Informatik, Techn. Univ. Chemnitz, June 2004
(PDF Document).
Torsten Mehlan, Wolfgang Rehm:: VIA2SISCI -- A new library that provides the VIA semantics for SCI connected clusters
In proceedings of 7th Workshop `Parallel Systems and Algorithms` (PASA'04) held in conjunction with ARCS'04, Augsburg, Germany March 23-26 2004
The paper is published within the Gesellschaft für Informatik 'Lecture Notes in Informatics', ISSN 1617-5468, ISBN 3-88579-370-9.
This paper is copyrighted by Gesellschaft für Informatik!
(PDF Document) (c) 2004 GI
Rene Grabner, Frank Mietke, Wolfgang Rehm:: An MPICH2 Channel Device Implementation over VAPI on InfiniBand
In proceedings of CAC'04, Workshop on Communication Architecture for Clusters held in conjunction with IPDPS 2004, April 26-30 2004, Santa Fe, New Mexico.
This paper is copyrighted by IEEE!
(PDF Document) (c) 2004 IEEE

Older Stuff

of 2003

of 2002

of 2001

of 2000

of 1999

of 1998

of 1997

of 1996

of 1995

of 1994

of 1993

Computer Architecture Technical Reports (CATs)

Robert Kullmann, Torsten Hoefler:: A short Performance Analysis of Abinit under different build environments
Computer Architecture Technical Report, RA-TR-2006-1, Chemnitz University of Technology,
Dept. of Computer Science, Jan 2006
(PDF Document)
Torsten Hoefler, Wolfgang Rehm:: A short Performance Analysis of Abinit on a Cluster System
Computer Architecture Technical Report, RA-TR-2005-2, Chemnitz University of Technology,
Dept. of Computer Science, July 2005
(PDF Document)
Lavinio Cerquetti:: A Brief Structural Analysis of the Open MPI PTL API (Draft)
Computer Architecture Technical Report, RA-TR-2005-1, Chemnitz University of Technology,
Dept. of Computer Science, January 2005, (PDF Document)
Christian Siebert:: One-sided Synchronization - A new Approach to Synchronization Primitives
Computer Architecture Technical Report, Chemnitz University of Technology, Dept. of Computer Science, July 2004
(PDF Document).
Torsten Hoefler:: Meta Analysis of Gigabit Ethernet over Copper Solutions for Cluster-Networking
Computer Architecture Technical Report, Fakultaet fuer Informatik, Techn. Univ. Chemnitz, June 2004
(PDF Document).
Mario Trams:: Feasibility of PACX--MPI for use in a Cluster--of--Clusters Environment
Computer Architecture Technical Report, Fakultaet fuer Informatik, Techn. Univ. Chemnitz, December 2002
(PDF Document).
A. Willert:: Examination of Cactus on a Cluster-System using WaveDemo
Computer Architecture Technical Report, Fakultät für Informatik, Techn. Univ. Chemnitz, September 2002
(PDF Document)
S. Garg, A. Willert, W. Rehm:: Evaluating Performance of Heterogeneous Clusters Using Matrix Inversion
Computer Architecture Technical Report, Fakultaet fuer Informatik, Techn. Univ. Chemnitz, August 2002
(PDF Document
R. Schmidt:: Der NetPIPE-Benchmark, Beitrag zum Seminar "Benchmarking" im WS1999/2000
Computer Architecture Technical Report, Fakultät für Informatik, Techn. Univ. Chemnitz, June 2000
(ps.gz Document)
Friedrich Seifert:: Der LogP-Benchmark, Beitrag zum Seminar "Benchmarking" im SS98
Computer Architecture Technical Report, Fakultät für Informatik, Techn. Univ. Chemnitz, October 1998
(ps.gz Document).
Mario Trams:: Determining the PCI performance with the HP Exercizer/Analyzer (E2920A)
Computer Architecture Technical Report, Fakultät für Informatik, Techn. Univ. Chemnitz, November 1997
(ps.gz Document).
Thomas Radke:: More Message Passing Performance with the Multithreaded MPICH device
Computer Architecture Technical Report, Fakultät für Informatik, Techn. Univ. Chemnitz, 1997
(ps.gz Document).
J. Werner, L. Grabowsky, T. Radke:: SCI-MPI - An optimized implementation of a MPI subset for SCI connected SMP-systems
Computer Architecture Technical Report, Fakultät für Informatik, Techn. Univ. Chemnitz, 1997
(ps.gz Document).
L. Grabowsky, T. Ermer, J. Werner:: Nutzung von MPI für parallele FEM-Systeme
Computer Architecture Technical Report, Fakultät für Informatik, Techn. Univ. Chemnitz, 1997
(ps.gz Document).
O. Langer:: Threadlokale Variablen - Ein Präprozessor
Computer Architecture Technical Report, Fakultät für Informatik, Techn. Univ. Chemnitz, 1997
(ps.gz Document).