Why use a fishing line when you have a net? an adaptive multicast data distribution protocol

Authors:
Jeremy R. Cooperstock;Steve Kotsopoulos
Affiliations:
Department of Electrical and Computer Engineering, University of Toronto;Department of Electrical and Computer Engineering, University of Toronto
Venue:
ATEC '96 Proceedings of the 1996 annual conference on USENIX Annual Technical Conference
Year:
1996

Citing 6
Cited 5

The V distributed system

Communications of the ACM
UNIX network programming

UNIX network programming
The process group approach to reliable distributed computing

Communications of the ACM
The Information Bus: an architecture for extensible distributed systems

SOSP '93 Proceedings of the fourteenth ACM symposium on Operating systems principles
Conversationbuilder: a collaborative erector set

Conversationbuilder: a collaborative erector set
Publishing: a reliable broadcast communication mechanism

SOSP '83 Proceedings of the ninth ACM symposium on Operating systems principles

Network text editor (NTE): A scalable shared text editor for the MBone

SIGCOMM '97 Proceedings of the ACM SIGCOMM '97 conference on Applications, technologies, architectures, and protocols for computer communication
New channels, old concerns: scalable and reliable data dissemination

EW 9 Proceedings of the 9th workshop on ACM SIGOPS European workshop: beyond the PC: new challenges for the operating system
MTCP: scalable TCP-like congestion control for reliable multicast

Computer Networks: The International Journal of Computer and Telecommunications Networking
Partition Cast - Modelling and Optimizing the Distribution of Large Data Sets in PC Clusters (Distinguished Paper)

Euro-Par '00 Proceedings from the 6th International Euro-Par Conference on Parallel Processing
Programming distributed systems with group IO

EUROMICRO-PDP'02 Proceedings of the 10th Euromicro conference on Parallel, distributed and network-based processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

The design and implementation of a system to provide reliable and efficient distribution of large quantities of data to many hosts on a local area network or internetwork is described. By exploiting the one-to-many transmission capabilities of multicast and broadcast, it is possible to transmit data to multiple hosts simultaneously, using less bandwidth and thus obtaining greater efficiency than repeated unicasting. Although performance measurements indicate the superiority of multicast, we dynamically select from available transmission modes so as to maximize efficiency and throughput while providing reliable delivery of data to all hosts. Our results demonstrate that file-distribution programs based on our protocol can benefit from a substantial speed-up over TCP-based programs such as rdist. For example, our system has been used to distribute a 133 Kbyte password file to 68 hosts in 20 seconds, whereas the equivalent rdist took 251 seconds.