Time-Optimal $k$-Server

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Frei, Fabian, Komm, Dennis, Stocker, Moritz, Whittington, Philip
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912264644198400
author Frei, Fabian
Komm, Dennis
Stocker, Moritz
Whittington, Philip
author_facet Frei, Fabian
Komm, Dennis
Stocker, Moritz
Whittington, Philip
contents The time-optimal $k$-server problem minimizes the time spent serving all requests instead of the distances traveled. We give a lower bound of $2k-1$ on the competitive ratio of any deterministic online algorithm for this problem, which coincides with the best known upper bound on the competitive ratio achieved by the work-function algorithm for the classical $k$-server problem. We provide further lower bounds of $k+1$ for all Euclidean spaces and $k$ for uniform metric spaces. For the latter, we give a matching $k$-competitive deterministic algorithm. Our most technical result, proven by applying Yao's principle to a suitable instance distribution on a specifically constructed metric space, is a lower bound of $k+\mathcal{O}(\log k)$ that holds even for randomized algorithms, which contrasts with the best known lower bound for the classical problem that remains polylogarithmic. With this paper, we hope to initiate a further study of this natural yet neglected problem.
format Preprint
id arxiv_https___arxiv_org_abs_2503_05589
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Time-Optimal $k$-Server
Frei, Fabian
Komm, Dennis
Stocker, Moritz
Whittington, Philip
Data Structures and Algorithms
The time-optimal $k$-server problem minimizes the time spent serving all requests instead of the distances traveled. We give a lower bound of $2k-1$ on the competitive ratio of any deterministic online algorithm for this problem, which coincides with the best known upper bound on the competitive ratio achieved by the work-function algorithm for the classical $k$-server problem. We provide further lower bounds of $k+1$ for all Euclidean spaces and $k$ for uniform metric spaces. For the latter, we give a matching $k$-competitive deterministic algorithm. Our most technical result, proven by applying Yao's principle to a suitable instance distribution on a specifically constructed metric space, is a lower bound of $k+\mathcal{O}(\log k)$ that holds even for randomized algorithms, which contrasts with the best known lower bound for the classical problem that remains polylogarithmic. With this paper, we hope to initiate a further study of this natural yet neglected problem.
title Time-Optimal $k$-Server
topic Data Structures and Algorithms
url https://arxiv.org/abs/2503.05589