Systems and Means of Informatics
2014, Volume 24, Issue 4, pp 135-156
METHODS OF OPTIMIZATION OF SNOOPING CACHE COHERENCE SYSTEMS
Abstract
During development of multiprocessor systems, an important task is maintenance of consistency (coherence) of data local cache processors. To ensure consistency of data, each processor in the system checks other processors on the presence of shared data in their cache by sending requests on network processors. However, a significant part of such requests is redundant, which increases network traffic and energy consumption when checking address tags. This article provides an overview of various methods of eliminating redundant requests in snoop-based coherence systems. The final section provides comparative characteristics of these methods and recommendations for the effective use of each method.
[+] References (7)
- Cantin, J., M. Lipasti, and J. Smith. 2005. Improving multiprocessor performance with coarse-grain coherence tracking. ISCA'05: 32nd Annual Symposium (International) on Computer Architecture Proceedings. Washington: IEEE Computer Society. 246-257.
- Moshovos, A., G. Memik, A. Choudhary, and B. Falsafi. 2001. JETTY: Filtering snoops for reduced energy consumption in SMP servers. HPCA'01: 7th Symposium (International) on High-Performance Computer Architecture Proceedings. Washington: IEEE Computer Society. 85-96.
- Salapura, V., M. Blumrich., and A. Gara. 2007. Improving the accuracy of snoop filtering using stream registers. MEDEA'07: 2007 Workshop on Memory Performance:
Dealing with Applications, Systems, and Architecture Proceedings. New York, ACM. 25-32.
- Blumrich, M. A., V. Salapura, and A. Gara. 2011. Exploring the architecture of a stream register-based snoop filter. Transactions on high-performance embedded architectures and compilers III. Springer. 93-114.
- Ranganathan, A., A. G. Bayrak, T. Kluter, P. Brisk, E. Charbon, and P. Ienne. 2012. Counting stream registers: An efficient and effective snoop filter architecture. 2012 Conference (International) on Embedded Computer Systems Proceedings. Washington: IEEE Computer Society. 120-127.
- Moshovos, A. 2005. RegionScout: Exploiting coarse grain sharing in snoop-based coherence. ISCA'05: 32nd Annual Symposium (International) on Computer Architecture Proceedings. Washington, IEEE Computer Society. 234-245.
- Kim, D., J. Ahn, J. Kim., and J. Huh. 2010. Subspace snooping: Filtering snoops with operating system support appears. PACT'10: 19th Conference (International) on Parallel Architectures and Compilation Techniques Proceedings. New York: ACM. 111-122.
[+] About this article
Title
METHODS OF OPTIMIZATION OF SNOOPING CACHE COHERENCE SYSTEMS
Journal
Systems and Means of Informatics
Volume 24, Issue 4, pp 135-156
Cover Date
2013-11-30
DOI
10.14357/08696527140409
Print ISSN
0869-6527
Publisher
Institute of Informatics Problems, Russian Academy of Sciences
Additional Links
Key words
multiprocessor systems; snoop-based coherence systems; page table; translation lookaside buffer
Authors
B. Z. Shmeilin 
Author Affiliations
Institute of Informatics Problems, Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation
|