Solving the at-most-once problem with nearly optimal effectiveness

  • Authors:
  • Sotirios Kentros;Aggelos Kiayias

  • Affiliations:
  • University of Connecticut, Storrs, CT, USA;University of Connecticut, Storrs, CT, USA

  • Venue:
  • Proceedings of the 30th annual ACM SIGACT-SIGOPS symposium on Principles of distributed computing
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

We present and analyze a wait-free deterministic algorithm for solving the at-most-once problem: how m fail-prone processes perform asynchronously n tasks at most once using shared memory. Our algorithmic strategy provides for the first time nearly optimal effectiveness, which is a measure that expresses the total number of tasks completed in the worst case. Our algorithm's effectiveness equals n-2m+2. This is up to an additive factor of m close to the known effectiveness lower bound n-m+1 and improves on the previously best known deterministic solutions that have effectiveness only n-log m o(n). We also present a work and space complexity analysis for suitable ranges of the algorithm parameters and demonstrate further that (i) we can achieve work O(nm log n log m) and simultaneously effectiveness of n-3m2-m+2, which is asymptotically optimal for any m=o(√n),(ii) we can achieve optimal work up to logarithmic factors Õ(n) and asymptotically optimal effectiveness whenever m=o(3√n).