Distributed Clustering in General Metrics via Coresets

Center-based clustering is a fundamental primitive for data analysis and is very challenging for large datasets. We developed coreset based space/round-efficient MapReduce algorithms to solve the k-center, k-median, and k-means variants in general metrics. Remarkably, the algorithms obliviously adapt to the doubling dimension of the metric space, and attain approximation ratios that can be made arbitrarily close to those achievable by the best known polynomial-time sequential approximations.