EXPLORING NON-TYPICAL MEMCACHE ARCHITECTURES FOR DECREASED LATENCY AND DISTRIBUTED NETWORK USAGE

Paul Talaga, Steve Chapin

2012

Abstract

Memcache is a distributed in-memory data store designed to reduce database load for web applications by caching frequently used data across multiple machines. In a distributed web serving environment applications rely on many network services to complete each request. While faster processors have lowered computation time and available network bandwidth has increased, signal propagation delay is a constant and will become a larger proportion of latency in the future. We explore how data-locality with Memcache can be exploited to reduce latency and minimize core network traffic. A model is developed to predict how alternate Memcache configurations would perform for specific applications followed by an evaluation using the MediaWiki opensource web application in a miniature web farm setting. Our results verified our model and we observed a 66% reduction in core network traffic and a 23% reduction in Memcache response time under certain network conditions.

References

  1. Aldinucci, M. and Torquati, M. (2004). Accelerating Apache Farms Through Ad-HOC Distributed Scalable Object Repository, volume 3149 of Lecture Notes in Computer Science. Springer Berlin / Heidelberg.
  2. Bailey, N. (2011). Frontpage - cassandra wiki.
  3. Borthakur, D. (2011). Hdfs architecture guide.
  4. Cloudkick (2011). Visual evidence of amazon ec2 network issues.
  5. Hennessy, J. L. and Patterson, D. A. (2006). Computer Architecture, Fourth Edition: A Quantitative Approach. Morgan Kaufmann Publishers Inc., San Francisco, CA, USA.
  6. KLab (2011). repcached - add data replication feature to memcached.
  7. Li, K. and Hudak, P. (1989). Memory coherence in shared virtual memory systems. ACM Trans. Comput. Syst., 7:321-359.
  8. mediawiki (2011). Mediawiki.
  9. Newman, S. (2011). Three latency anomalies.
  10. Nithya Sampathkumar, Muralidhar Krishnaprasad, A. N. (2009). Introduction to caching with windows server appfabric.
  11. rsumbaly (2011). Voldemort topology awareness capability.
  12. RuggedCom (2011). Latency on a switched ethernet network.
  13. Saab, P. (2008). Scaling memcached at facebook.
  14. Tanenbaum, A. S. and Steen, M. V. (2001). Distributed Systems: Principles and Paradigms. Prentice Hall PTR, Upper Saddle River, NJ, USA, 1st edition.
  15. Tanenbaum, A. S., Tanenbaum, A. S., Kaashoek, M. F., Kaashoek, M. F., Bal, H. E., and Bal, H. E. (1994). Using broadcasting to implement distributed shared memory efficiently. In Readings in Distributed Computing Systems, pages 387-408. IEEE Computer Society Press.
  16. Terracotta (2011). Ehcache documentation cache-topologies.
  17. Voldemort, P. (2011). Project voldemort.
Download


Paper Citation


in Harvard Style

Talaga P. and Chapin S. (2012). EXPLORING NON-TYPICAL MEMCACHE ARCHITECTURES FOR DECREASED LATENCY AND DISTRIBUTED NETWORK USAGE . In Proceedings of the 8th International Conference on Web Information Systems and Technologies - Volume 1: WEBIST, ISBN 978-989-8565-08-2, pages 36-46. DOI: 10.5220/0003933200360046


in Bibtex Style

@conference{webist12,
author={Paul Talaga and Steve Chapin},
title={EXPLORING NON-TYPICAL MEMCACHE ARCHITECTURES FOR DECREASED LATENCY AND DISTRIBUTED NETWORK USAGE},
booktitle={Proceedings of the 8th International Conference on Web Information Systems and Technologies - Volume 1: WEBIST,},
year={2012},
pages={36-46},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0003933200360046},
isbn={978-989-8565-08-2},
}


in EndNote Style

TY - CONF
JO - Proceedings of the 8th International Conference on Web Information Systems and Technologies - Volume 1: WEBIST,
TI - EXPLORING NON-TYPICAL MEMCACHE ARCHITECTURES FOR DECREASED LATENCY AND DISTRIBUTED NETWORK USAGE
SN - 978-989-8565-08-2
AU - Talaga P.
AU - Chapin S.
PY - 2012
SP - 36
EP - 46
DO - 10.5220/0003933200360046