|
| This Article | ||
| ||
| Share | ||
| Bibliographic References | ||
| Add to: | ||
 Digg  Furl  Spurl  Blink  Simpy  Google  Del.icio.us  Y!MyWeb | ||
| Search | ||
| ||
| ASCII Text | x | ||
| Dionysios Kostoulas, Dimitrios Psaltoulis, Indranil Gupta, Ken Birman, Al Demers, "Decentralized Schemes for Size Estimation in Large and Dynamic Groups," Network Computing and Applications, IEEE International Symposium on, pp. 41-48, Fourth IEEE International Symposium on Network Computing and Applications, 2005. | |||
| BibTex | x | ||
| @article{ 10.1109/NCA.2005.15, author = {Dionysios Kostoulas and Dimitrios Psaltoulis and Indranil Gupta and Ken Birman and Al Demers}, title = {Decentralized Schemes for Size Estimation in Large and Dynamic Groups}, journal ={Network Computing and Applications, IEEE International Symposium on}, volume = {0}, year = {2005}, isbn = {0-7695-2326-9}, pages = {41-48}, doi = {http://doi.ieeecomputersociety.org/10.1109/NCA.2005.15}, publisher = {IEEE Computer Society}, address = {Los Alamitos, CA, USA}, } | |||
| RefWorks Procite/RefMan/Endnote | x | ||
| TY - CONF JO - Network Computing and Applications, IEEE International Symposium on TI - Decentralized Schemes for Size Estimation in Large and Dynamic Groups SN - 0-7695-2326-9 SP41 EP48 A1 - Dionysios Kostoulas, A1 - Dimitrios Psaltoulis, A1 - Indranil Gupta, A1 - Ken Birman, A1 - Al Demers, PY - 2005 KW - null VL - 0 JA - Network Computing and Applications, IEEE International Symposium on ER - | |||
Large-scale and dynamically changing distributed systems such as the Grid, peer-to-peer overlays, etc., need to collect several kinds of global statistics in a decentralized manner. In this paper, we tackle a specific statistic collection problem called Group Size Estimation, for estimating the number of non-faulty processes present in the global group at any given point of time. We present two new decentralized algorithms for estimation in dynamic groups, analyze the algorithms, and experimentally evaluate them using real-life traces. One scheme is active: it spreads a gossip into the overlay first, and then samples the receipt times of this gossip at different processes. The second scheme is passive: it measures the density of processes when their identifiers are hashed into a real interval. Both schemes have low latency, scalable perprocess overheads, and provide high levels of probabilistic accuracy for the estimate. They are implemented as part of a size estimation utility called PeerCounter that can be incorporated modularly into standard peer-to-peer overlays. We present experimental results from both the simulations and PeerCounter, running on a cluster of 33 Linux servers.