Approximating the average response time in broadcast scheduling

  • Authors:
  • Nikhil Bansal;Moses Charikar;Sanjeev Khanna;Joseph (Seffi) Naor

  • Affiliations:
  • IBM T.J. Watson Research Center, Yorktown Heights, NY;Princeton University, Princeton, NJ;University of Pennsylvania, Philadelphia PA;Technion, Haifa, Israel

  • Venue:
  • SODA '05 Proceedings of the sixteenth annual ACM-SIAM symposium on Discrete algorithms
  • Year:
  • 2005

Quantified Score

Hi-index 0.00

Visualization

Abstract

We consider the problem of approximating the minimum average response time in on-demand data broadcasting systems. The best approximation factors known for this problem involve resource augmentation. We provide the first non-trivial approximation factors in the absence of resource augmentation, achieving an additive O(√n)-approximation, where n is the number of distinct pages. Our result can be extended, for any ε 0, to a (1 + ε)-speed, additive O(1/ε)-approximation algorithm. Prior to our work, no non-trivial approximation factor was known for the case of ε