Fast causal multicast

  • Authors:
  • Pat Stephenson;Kenneth Birman

  • Affiliations:
  • Cornell University, Ithaca NY;Cornell University, Ithaca NY

  • Venue:
  • EW 4 Proceedings of the 4th workshop on ACM SIGOPS European workshop
  • Year:
  • 1990

Quantified Score

Hi-index 0.00

Visualization

Abstract

We begin by outlining a new protocol that efficiently implements a reliable, causally ordered multicast primitive and is easily extended into a totally ordered one. Since measurements show that the dominant cost of this protocol is message transport, the design of a lower level multicast transport protocol is discussed. The overall scheme scales with bounded overhead. Our first conclusion is that systems such as ISIS can achieve performance competitive with the best existing multicast facilities - a finding contradicting the widespread concern that fault-tolerance may be unacceptably costly. Our second conclusion is that the paradigm of multicast transport is extremely useful for constructing fault-tolerant applications. Our final conclusion is that the framework for fault-tolerant programming provided by ISIS is very useful in the design and construction of these extensions to the basic system.