Active and passive techniques for group size estimation in large-scale and dynamic distributed systems

This paper presents two solutions to a distributed statistic collection problem, called Group Size Estimation. These algorithms are intended for large-scale and dynamic distributed systems such as Grids, peer-to-peer overlays, etc. Each algorithm estimates (both in a one-shot and continuous manner)...

Full description

Saved in:
Bibliographic Details
Published inThe Journal of systems and software Vol. 80; no. 10; pp. 1639 - 1658
Main Authors Kostoulas, Dionysios, Psaltoulis, Dimitrios, Gupta, Indranil, Birman, Kenneth P., Demers, Alan J.
Format Journal Article
LanguageEnglish
Published New York Elsevier Inc 01.10.2007
Elsevier Sequoia S.A
Subjects
Algorithm design and analysis
Algorithms
Computer simulation
Distributed processing
Epidemics
Estimating techniques
Group size estimation
Large-scale distributed systems
Probabilistic protocols
Studies
Algorithm design and analysis
Group size estimation
Epidemics
Large-scale distributed systems
Probabilistic protocols
Online AccessGet full text

Cover

More Information
Summary:This paper presents two solutions to a distributed statistic collection problem, called Group Size Estimation. These algorithms are intended for large-scale and dynamic distributed systems such as Grids, peer-to-peer overlays, etc. Each algorithm estimates (both in a one-shot and continuous manner) the number of non-faulty processes present in the global group. The first active scheme samples receipt times of gossip messages, while the second passive scheme calculates the density of process identifiers when hashed to a real interval. Our analysis, trace-driven simulation and deployment on a 33-node Linux cluster study and compare the latencies, scalability, and accuracy of these schemes.
ISSN:0164-1212
1873-1228
DOI:10.1016/j.jss.2007.01.014