Reducing data access latency in SDSM systems using runtime optimizations
Authors: 
Javier Bueno, Xavier Martorell, Juan José Costa, Toni Cortés, Eduard Ayguadé, Guansong Zhang, Christopher Barton, Raul SilveraAuthors Info & Claims
Javier Bueno, Xavier Martorell, Juan José Costa, Toni Cortés, Eduard Ayguadé, Guansong Zhang, Christopher Barton, Raul SilveraAuthors Info & ClaimsPages 160 - 173
Abstract
Software Distributed Shared Memory (SDSM) systems offer a convenient way to run applications developed for shared memory systems on distributed systems with no changes to them. However, since SDSM systems add an extra layer of abstraction to the memory hierarchy, applications may suffer performance problems when running on top of them.
Our main research interest is to develop a set of compiler and runtime system techniques that widen the range of applications that can efficiently run on SDSM systems. Currently we are targeting OpenMP applications due to the ease of use this programming model provides.
In this paper we show the performance of a set of regular applications that perform well on our SDSM system. They were adapted from OpenCL codes provided by ATI, and re-written in OpenMP. When trying to exploit more complex applications with different data access patterns, we find more difficulties from a DSM system. As an example, we show the performance evaluation of the NAS MG benchmark, and two techniques we have developed to improve its data locality.
Our SDSM infrastructure is composed of NanosDSM, an everything-shared SDSM developed at the Technical University of Catalonia (UPC) and the Barcelona Supercomputing Center (BSC), and the IBM XL SMP Runtime to allow the execution of the OpenMP applications.
References
[1]
NANOS project overview. http://nanos.ac.upc.edu.
[2]
J. Balart, A. Duran, M. González, X. Martorell, E. Ayguadé, and J. Labarta. Nanos mercurium: a research compiler for openmp. In Sixth European Workshop on OpenMP, Stockholm, Sweden, 2004.
[3]
C. Amza, A. L. Cox, S. Dwarkadas, P. Keleher, H. Lu, R. Rajamony, W. Yu, W. Zwaenepoel. TreadMarks: Shared Memory Computing on Networks of Workstations. IEEE COmputer 29 (2), pages 18--28, 1996.
[4]
JJ Costa, T. Cortes, X. Martorell, E. Ayguade, and J. Labarta. Running OpenMP applications efficiently on an everything-shared SDSM. Journal of Parallel and Distributed Computing, 66(5):647--658, 2006.
[5]
E. Speight, J. K. Bennett. Brazos: A Third Generation DSM System. In Proc. of the USENIX Windows NT Workshop, 1997.
[6]
OpenMP Forum. OpenMP: A Proposed Industry Standard API for Shared Memory Programming. 1997.
[7]
J. P. Hoelflinger. Extending OpenMP to Clusters. 2006.
[8]
L. Whately, R. Pinto, R. Bianchini, C. L. Amorim. Adaptative Techniques for Home-Based Software DSMs. 13th Symposium on Computer Architecture and High Performance Computing, 2001.
[9]
M. Hess, G. Jost, M. Müller, R. Rühle. Experiences using OpenMP based on Compiler Directed Software DSM on a PC Cluster. In Workshop on OpenMP Applications and Tools (WOMPAT'02, 2002.
[10]
M. Sato, S. Satoh, K. Kusano, Y. Tanaka. Design of an OpenMP Compiler for an SMP Cluster. In EWOMP'99, pages 32--39, 1999.
[11]
X. Martorell, J. Labarta, J. I. Navarro, and E. Ayguadé. A library implementation of the nano-threads programming model. In Lecture Notes in Computer Science, Euro-Par'96, pages 644--649, Springer-Verlag, August 1996.
[12]
W. Hu, W. Shi, Z. Tang. JIAJIA: An SVM System Bbased on A New Cache Coherence Protocol. In Proceedings of the High Performance Computing and Networking (HPCN'99), pages 463--472, Springer, Amsterdam, Netherlands, 1999.
[13]
Y. C. Hu, H. Lu, A. L. Cox, W. Zwaenepoel. OpenMP for Networks of SMPs. Journal of Parallel and Distributed Computing 60 (12), pages 1512--1530, 2000.
[14]
Y. Kee, J. Kim, S. Ha. ParADE: An OpenMP Programming Environment for SMP Cluster Systems. In Supercomputing 2003 (SC'03), 2003.
Index Terms
- Reducing data access latency in SDSM systems using runtime optimizations
Recommendations
P3DC: Reducing DRAM Cache Hit Latency by Hybrid Mappings
AbstractDie-stacked dynamic random access memory (DRAM) caches are increasingly advocated to bridge the performance gap between the on-chip cache and the main memory. To fully realize their potential, it is essential to improve DRAM cache hit rate and ...
Codesign of NoC and Cache Organization for Reducing Access Latency in Chip Multiprocessors
Reducing data access latency is vital to achieving performance improvements in computing. For chip multiprocessors (CMPs), data access latency depends on the organization of the memory hierarchy, the on-chip interconnect, and the running workload. ...
Comments
Information & Contributors
Information
Published In
November 2010
482 pages
- Conference Chairs:
- Joanna Ng,
- Christian Couturier,
- Editors:
- Hausi A. Müller,
- Arthur Ryman,
- Anatol W. Kark,
- Program Chairs:
- Hausi A. Müller,
- Arthur Ryman
Publisher
IBM Corp.
United States
Publication History
Published: 01 November 2010
Qualifiers
- Research-article
Funding Sources
Conference
CASCON '10
CASCON '10: Center for Advanced Studies on Collaborative Research
November 1 - 4, 2010
Ontario, Toronto, Canada
Acceptance Rates
Overall Acceptance Rate 24 of 90 submissions, 27%
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- 0Total Citations
- 77Total Downloads
- Downloads (Last 12 months)0
- Downloads (Last 6 weeks)0
Reflects downloads up to 29 Jan 2025
Other Metrics
Citations
View Options
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Sign in