Geographic quorum system approximations

Quorum systems are a mechanism for obtaining fault-tolerance and efficient distributed systems. We consider geographic quorum systems; a geographic quorum system is a partition of a set Χ of sites in the plane (representing servers) into quorums (i.e., clusters) of size k. The distance between a point p and a cluster C is the Euclidean distance between p and the site in C that is the farthest from p. We present a near linear time constant-factor approximation algorithm for partitioning Χ into clusters, such that the maximal distance between a point in the underlying region and its closest cluster is minimized. Next, we describe a data structure for answering (approximately) nearest-neighbor queries on such a clustering. Finally, we study the problem of partitioning into clusters with an additional load-balancing requirement.