Title
Online Chasing Problems for Regular n-Gons
Abstract
We consider a server location problem with only one server to move. If each request must be served on the exact position, there is no choice for the online player and the problem is trivial. In this paper we assume that a request is given as a region and that the service can be done anywhere inside the region. Namely, for each request an online algorithm chooses an arbitrary point in the region and moves the server there. Our main result shows that if the region is a regular n-gon, the competitive ratio of the greedy algorithm is 1/sin pi/2n for odd n and 1/sin pi/n for even n. Especially for a square region, the greedy algorithm turns out to be optimal.
Year
DOI
Venue
2007
10.1109/RIVF.2007.369133
RIVF
Keywords
Field
DocType
cellular neural networks,k server problem,competitive ratio,tellurium,greedy algorithms,online algorithm,greedy algorithm,broadcasting,queueing theory,space exploration,shape,computational geometry,layout
Online algorithm,Broadcasting,Combinatorics,Computational geometry,K-server problem,Greedy algorithm,Queueing theory,Mathematics,Competitive analysis
Conference
Citations 
PageRank 
References 
0
0.34
7
Authors
3
Name
Order
Citations
PageRank
H. Fujiwara17314.23
Kazuo Iwama21400153.38
Kouki Yonezawa3174.05