On runtime adaptation of application-layer multicast protocol parameters

Authors:
Christian Hübsch;Christoph P. Mayer;Oliver P. Waldhorst
Affiliations:
Institute of Telematics, Karlsruhe Institute of Technology, Germany;Institute of Telematics, Karlsruhe Institute of Technology, Germany;Institute of Telematics, Karlsruhe Institute of Technology, Germany
Venue:
EUNICE'10 Proceedings of the 16th EUNICE/IFIP WG 6.6 conference on Networked services and applications: engineering, control and management
Year:
2010

Citing 5
Cited 0

Scalable application layer multicast

Proceedings of the 2002 conference on Applications, technologies, architectures, and protocols for computer communications
Bandwidth-efficient management of DHT routing tables

NSDI'05 Proceedings of the 2nd conference on Symposium on Networked Systems Design & Implementation - Volume 2
MOSAIC: unified declarative platform for dynamic overlay composition

CoNEXT '08 Proceedings of the 2008 ACM CoNEXT Conference
T-Man: gossip-based overlay topology management

ESOA'05 Proceedings of the Third international conference on Engineering Self-Organising Systems
A Survey of Application-Layer Multicast Protocols

IEEE Communications Surveys & Tutorials

Quantified Score

Hi-index	0.00

Visualization

Abstract

Reasonable choice of protocol parameters is crucial for the successful deployment of overlay networks fulfilling given service quality requirements in next generation networks. Unfortunately, changing network conditions, as well as changing application and user requirements may invalidate an initial parameter choice during the lifetime of an overlay. To this end, runtime adaptation of protocol parameters seems to be a promising solution--however, it is not clear if protocol parameters can be adjusted dynamically at runtime in a distributed setting. In this paper, we show--using the NICE application layer multicast protocol as an example--that runtime adaptation of protocol parameters is indeed feasible. We propose an algorithm for adapting the NICE clustersize parameter k dynamically at runtime and discuss the impact on service quality. Our simulations show that runtime adaptation of NICE protocol parameters is promising for service improvement and that data latencies can be optimized by up to 25% without increasing the overhead significantly for most of the nodes.