Home

About

Advanced Search

Browse by Discipline

Scientific Societies

E-print Alerts

Add E-prints

E-print Network
FAQHELPSITE MAPCONTACT US


  Advanced Search  

 
Ecient Server Replication and Client Re-direction for Multicast Services
 

Summary: E∆cient Server Replication and Client Re-direction for
Multicast Services
Zongming Fei a and Mostafa Ammar b and Ellen Zegura b
a Department of Computer Science, University of Kentucky,
773 Anderson Hall, Lexington, KY 40506-0046
b Networking and Telecommunications Group, College of Computing,
Georgia Tech, Atlanta, GA 30332-0280
ABSTRACT
Server replication and multicasting are well-established techniques for increasing the capacity of a networked service
and improving client performance. In this paper, we consider the combination of these two techniques. Speci cally,
we investigate the problem of selecting amongst rate-adaptive multicast servers, which adjust their sending rate based
on network conditions and/or feedback from clients. E ective server rate adaptation can lead to e∆cient utilization
of network resources and performance improvement perceived by clients. In this study of adaptive multicast server
selection, we explore some fundamental issues and investigate the implications of di erent selection strategies on the
performance perceived by clients. We rst formulate several optimization problems based on di erent performance
measures. We prove that the general problem is NP-hard and then present a special case with an optimal polynomial-
time solution. We then design several heuristics for the general problem. We investigate the performance of the
heuristics through simulation and show that our heuristics can improve the performance perceived by clients over
other strategies.
Keywords: Multicast, Server Selection, Replication, Adaptive Server

  

Source: Ammar, Mostafa - College of Computing, Georgia Institute of Technology
Fei, Zongming - Department of Computer Science, University of Kentucky
Zegura, Ellen W. - College of Computing, Georgia Institute of Technology

 

Collections: Computer Technologies and Information Sciences