We study the impact of additional information on the hardness of the k-server problem on different metric spaces. To this end, we consider the well-known model of computing with advice. In particular, we design an algorithm for the d-dimensional Euclidean space, which generalizes a known result for the Euclidean plane. As another relevant setting, we investigate a metric space with positive curvature; in particular, the sphere. Both algorithms have constant strict competitive ratios while reading a constant number of advice bits with every request, independent of the number k of servers, and solely depending on parameters of the underlying metric structure.
CITATION STYLE
Burjons, E., Komm, D., & Schöngens, M. (2018). The k-server problem with advice in d dimensions and on the sphere. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 10706 LNCS, pp. 396–409). Springer Verlag. https://doi.org/10.1007/978-3-319-73117-9_28
Mendeley helps you to discover research relevant for your work.