A lower bound on the competitivity of memoryless algorithms for a generalization of the CNN problem

Authors:
Yves F. Verhoeven
Affiliations:
Université Paris-Sud, LRI, Bâtiment, Orsay Cedex, France
Venue:
Theoretical Computer Science
Year:
2006

Citing 9
Cited 0

Competitive algorithms for server problems

Journal of Algorithms
An optimal on-line algorithm for K-servers on trees

SIAM Journal on Computing
New results on server problems

SIAM Journal on Discrete Mathematics
On the k-server conjecture

Journal of the ACM (JACM)
Online computation and competitive analysis

Online computation and competitive analysis
The harmonic k-server algorithm is competitive

Journal of the ACM (JACM)
The weighted 2-server problem

Theoretical Computer Science - Special issue: Online algorithms in memoriam, Steve Seiden
The CNN problem and other k-server variants

Theoretical Computer Science - Special issue: Online algorithms in memoriam, Steve Seiden
A competitive algorithm for the general 2-server problem

ICALP'03 Proceedings of the 30th international conference on Automata, languages and programming

Quantified Score

Hi-index	5.23

Visualization

Abstract

We consider the CNN problem in arbitrary dimension, and over any metric space containing the integers. We prove that, in every dimension at least 2, no memoryless online algorithm can achieve a constant competitive ratio, under a weak symmetry constraint on the algorithm. This generalizes in several aspects the lower bounds obtained by Koutsoupias and Taylor [The CNN Problem and other k-server variants, Theoret. Comput. Sci. 324 (2004) 347-359] for the original problem. The proof consists in the analysis of carefully selected random walks, which appear naturally in the framework of memoryless algorithms.