A dynamic application-driven data communication strategy
Pages 146 - 153
Abstract
The use of semi-Lagrangian formulations in numerical weather predication models (NWP) allows for an increase in time step size. Use of this method can increase performance of these models. However, on parallel architectures, communication between processors can become a huge bottleneck, limiting speedup. Furthermore, the communication pattern is dependent on the application's execution. We discus a novel strategy, called Halo on Demand, which dynamically drives the communication between the processors by examining the content of the data at runtime in order to reduce communication costs. With an extensive performance analysis of the execution of the model we show that our strategy can decrease communication time and thus decrease total execution time.
References
[1]
Siegfried Benkner, Kamran Sanjari, Viera Sipkova, and Bob Velkov. Parallelizing irregular applications with the vienna hpf+ compiler vfc. In Proceedings HPCN'98 (HPF+workshop), LNCS 1401, pages 797--808, Amsterdam, The Netherlands, April 1998. Springer Verlag.
[2]
Raoul A. F. Bhoudjang, Kees Verstoep, Tim Rühl, Henri E. Bal, and RutgerF. H. Hofman. Evaluating design alternatives for reliable communication on high-speed networks. In Proceedings of 9th International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS IX), pages 71--81, Cambridge, MA, USA, November 2000. ACM Press.
[3]
Hirlam home page. http://www.knmi.nl/hirlam.
[4]
The Distributed Asci Supercomputer home page. http://www.cs.vu.nl/das2/.
[5]
High performance fortran forum (hpff). http://www.crpc.rice.edu/HPFF/.
[6]
Myricom. http://www.myricom.com.
[7]
Performance of mpich-gm 1.2.4.8a on myrinet-2000. http://www.myricom.com/myrinet/performance/MPICH-GM/index.html.
[8]
MPICH-A Portable Implementation of MPI. http://www-unix.mcs.anl.gov/mpi/mpich/.
[9]
C. D. Polychronopolous and D. J. Kuck. Guided self-scheduling: A practical scheduling scheme for parallel supercomputers. IEEE Transactions on Computers, 36(12):1425--1439, December 1987.
[10]
The Beowulf project. http://www.beowulf.org.
[11]
J. Pudykiewicz and A. Staniforth. Some properties and comparative performance of the semi-langrangian method of robert in the solution of the advection-diffusion equation. Atmosphere-Ocean, 22(3):282--308, 1984.
[12]
M. J. Quinn. Parallel Computing, theory and practice. McGraw-Hill, 1994.
[13]
Joel H. Saltz, Ravi Mirchandaney, and Kay Crowley. Run-time parallelization and scheduling of loops. IEEE Transactions on Computers, 40(5):603--612, May 1991.
[14]
Andrew Staniforth and Jean Côté. Semi-langrangian integration schemes for atmospheric models -- a review. Monthly Weather Review, 119:2206--2223, September 1991.
[15]
Paul van der Mark, Lex Wolters, and Gerard Cats. Automatic code-generation for large scale applications. In Proceedings of the Tenth International Workshop on Compilers for Parallel Computers (CPC03), pages 157--164, Amsterdam, Netherlands, January 2003.
[16]
Paul van der Mark, Lex Wolters, and Gerard Cats. Code generation for semi-lagrangian formulation. In Proceedings of the Second International Symposium on Parallel and Distributed Computing (ISPDC'03), pages 266--273, Ljubljana, Slovenia, October 2003. IEEE Computer Society Press.
[17]
Paul van der Mark, Lex Wolters, and Gerard Cats. Semi-lagrangian formulations with automatic code generation for environmental modeling. In Proceedings of the 19th ACM Symposium on Applied Computing (SAC'04), pages 229--234, Nicosia, Cyprus, March 2004. ACM Press.
[18]
R. A. van Engelen. Ctadel: A Generator of Efficient Numerical Codes. PhD thesis, Universiteit Leiden, 1998.
Index Terms
- A dynamic application-driven data communication strategy
Recommendations
Dynamic branch prediction and control speculation
Branch prediction schemes have become an integral part of today's superscalar processors. They are one of the key issues in enhancing the performance of processors. Pipeline stalls due to conditional branches are one of the most significant impediments ...
Exploration of the Performance of a Data Mining Application via Hardware Based Monitoring
The performance of software on modern architectures has grown more and more difficult to predict and analyze, as modern microprocessors have grown more complex. The execution of a program now entails the complex interaction of code, compiler and ...
uops.info: Characterizing Latency, Throughput, and Port Usage of Instructions on Intel Microarchitectures
ASPLOS '19: Proceedings of the Twenty-Fourth International Conference on Architectural Support for Programming Languages and Operating SystemsModern microarchitectures are some of the world's most complex man-made systems. As a consequence, it is increasingly difficult to predict, explain, let alone optimize the performance of software running on such microarchitectures. As a basis for ...
Comments
Information & Contributors
Information
Published In
June 2004
360 pages
ISBN:1581138393
DOI:10.1145/1006209
- General Chair:
- Paul Feautrier,
- Program Chairs:
- James Goodman,
- André Seznec
Copyright © 2004 ACM.
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected]
Sponsors
Publisher
Association for Computing Machinery
New York, NY, United States
Publication History
Published: 26 June 2004
Check for updates
Author Tags
Qualifiers
- Article
Conference
Acceptance Rates
Overall Acceptance Rate 629 of 2,180 submissions, 29%
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- 0Total Citations
- 283Total Downloads
- Downloads (Last 12 months)3
- Downloads (Last 6 weeks)0
Reflects downloads up to 10 Aug 2024
Other Metrics
Citations
View Options
Get Access
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Sign in