TY - GEN
T1 - Decentralized schemes for size estimation in large and dynamic groups
AU - Kostoulas, Dionysios
AU - Psaltoulis, Dimitrios
AU - Gupta, Indranil
AU - Birman, Ken
AU - Demers, Al
PY - 2005
Y1 - 2005
N2 - 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.
AB - 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.
UR - http://www.scopus.com/inward/record.url?scp=33846296623&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=33846296623&partnerID=8YFLogxK
U2 - 10.1109/NCA.2005.15
DO - 10.1109/NCA.2005.15
M3 - Conference contribution
AN - SCOPUS:33846296623
SN - 0769523269
SN - 9780769523262
T3 - Proceedings - Fourth IEEE International Symposium on Network Computing and Applications, NCA 2005
SP - 41
EP - 48
BT - Proceedings - Fourth IEEE International Symposium on Network Computing and Applications, NCA 2005
T2 - 4th IEEE International Symposium on Network Computing and Applications, NCA 2005
Y2 - 27 July 2005 through 29 July 2005
ER -