Shared buffering in nonblocking copy networks

  • Manju V. Hegde
  • , P. S. Min
  • , A. Chandra
  • , H. Saidi

Research output: Contribution to conferencePaperpeer-review

Abstract

This paper investigates the performance of nonblocking copy networks employing the shared buffering scheme. We use the technique of tagged Markov chains to derive the stationary distributions for the occupancy of the buffers. Based on these stationary distributions, delay, throughput, and packet loss probability are calculated as critical performance measures of the copy network. We also carry out analyses for variations in the copying policy such as fanout splitting wherein a copy request from a single packet may be completed over two time slots, and fixed channel priority wherein there exists a certain priority among input channels.

Original languageEnglish
StatePublished - 1994
EventProceedings of the 1994 IEEE International Symposium on Information Theory - Trodheim, Norw
Duration: Jun 27 1994Jul 1 1994

Conference

ConferenceProceedings of the 1994 IEEE International Symposium on Information Theory
CityTrodheim, Norw
Period06/27/9407/1/94

Fingerprint

Dive into the research topics of 'Shared buffering in nonblocking copy networks'. Together they form a unique fingerprint.

Cite this