ScienceDirect® Home Skip Main Navigation Links
You have guest access to ScienceDirect. Find out more.
 
Home
Browse
My Settings
Alerts
Help
 Quick Search
 Search tips (Opens new window)
    Clear all fields    
Performance Evaluation
Volume 60, Issues 1-4, May 2005, Pages 31-50
Performance Modeling and Evaluation of High-Performance Parallel and Distributed Systems
 
Font Size: Decrease Font Size  Increase Font Size
 Abstract - selected
Article
Purchase PDF (632 K)

Article Toolbox
  E-mail Article   
  Add to my Quick Links   
Bookmark and share in 2collab (opens in new window)
Request permission to reuse this article
  Cited By in Scopus (0)
 
 
 
Related Articles in ScienceDirect
View More Related Articles
 
View Record in Scopus
 
doi:10.1016/j.peva.2004.10.003    
How to Cite or Link Using DOI (Opens New Window)

Copyright © 2004 Elsevier B.V. All rights reserved.

Simulation as a tool for optimizing memory accesses on NUMA machines

Purchase the full-text article



References and further reading may be available for this article. To view references and further reading you must purchase this article.

Jie Taoa, Corresponding Author Contact Information, E-mail The Corresponding Author, Martin Schulzb, 1, E-mail The Corresponding Author and Wolfgang Karla, E-mail The Corresponding Author

aInstitut für Rechnerentwurf und Fehlertoleranz, Universität Karlsruhe, Kaiserstr. 12, 76128 Karlsruhe, Germany

bSchool of Electrical and Computer Engineering, Cornell University, Ithaca, NY 14853, USA


Available online 7 December 2004.

Abstract

Due to the inherent non-uniformity in the memory system, programmers and users of non-uniform memory access (NUMA) machines have to take special care of the memory performance of their applications. This paper discusses a variety of potential improvements with respect to cache misses, cache invalidations, and inter-node communication. This study is based on the simulation tool SIMT, which models the memory hierarchy in detail and is capable of providing complete, accurate information about all dynamic memory references. This information can be used to analyze the memory access behavior of applications and thereby forms the basis for any optimization with respect to memory accesses.

Keywords: Simulation; Memory performance; NUMA machine; Locality optimization

Article Outline

1. Introduction
2. Performance challenges on NUMA systems
3. SIMT: an evaluation platform for multiprocessor systems
3.1. Simulation infrastructure
3.2. Architecture models
3.3. Cache and cache coherence protocols
3.4. Data allocation policies
3.5. Data migration
3.6. Data collection using a monitor simulator
3.7. Visualization of the performance data
3.8. Accuracy of SIMT
3.9. Summary
4. Sample memory optimization using SIMT
4.1. Cache locality optimization
4.2. Cache coherence protocols
4.3. Data distribution schemes
4.4. Migration impact
4.5. Summary
5. Related work
6. Conclusion
References












Corresponding Author Contact InformationCorresponding author. Tel.: +49 721 6086048; fax: +49 721 6083962.
1 Tel.: +1 607 2554997; fax: +1 607 2559072.

Performance Evaluation
Volume 60, Issues 1-4, May 2005, Pages 31-50
Performance Modeling and Evaluation of High-Performance Parallel and Distributed Systems
 
Home
Browse
My Settings
Alerts
Help
Elsevier.com (Opens new window)
About ScienceDirect  |  Contact Us  |  Information for Advertisers  |  Terms & Conditions  |  Privacy Policy
Copyright © 2008 Elsevier B.V. All rights reserved. ScienceDirect® is a registered trademark of Elsevier B.V.