Communication Efficient Algorithms for Top-k Selection Problems

TL;DR

This paper introduces scalable parallel algorithms for top-k selection and related problems, achieving low communication costs and efficient data redistribution in distributed systems.

Contribution

It presents novel parallel algorithms with sublinear communication for top-k selection, frequent object detection, and sum aggregation, including data redistribution techniques.

Findings

Algorithms have sublinear per-processor communication volume.

Efficient data redistribution minimizes communication overhead.

Applicable to various relevance notions and dynamic data.

Abstract

We present scalable parallel algorithms with sublinear per-processor communication volume and low latency for several fundamental problems related to finding the most relevant elements in a set, for various notions of relevance: We begin with the classical selection problem with unsorted input. We present generalizations with locally sorted inputs, dynamic content (bulk-parallel priority queues), and multiple criteria. Then we move on to finding frequent objects and top-k sum aggregation. Since it is unavoidable that the output of these algorithms might be unevenly distributed over the processors, we also explain how to redistribute this data with minimal communication.