Year: 2017
Type: article
Abstract: We study the problem of computing conjunctive queries over large databases on parallel architectures without shared storage. Using the structure of such a queryqand the skew in the data, we study tradeoffs between the number of processors, the number of rounds of communication, and the per-processorload—the number of bits each processor can send or can receive in a single round—that are required to computeq. Since each processor must store its received bits, the load is at most the number of bits of storage per processor. When the data are free of skew, we obtain essentially tight upper and lower bounds for one round algorithms, and we show how the bounds degrade when there is skew in the data. In the case of skewed data, we show how to improve the algorithms when approximate degrees of the (necessarily small number of) heavy-hitter elements are available, obtaining essentially optimal algorithms for queries such as skewed simple joins and skewed triangle join queries. For queries that we identify astreelike, we also prove nearly matching upper and lower bounds for multi-round algorithms for a natural class of skew-free databases. One consequence of these latter lower bounds is that for any ε > 0, usingpprocessors to compute the connected components of a graph, or to output the path, if any, between a specified pair of vertices of a graph withmedges and per-processor load that isO(m/p1−ε) requires Ω(logp) rounds of communication. Our upper bounds are given by simple structured algorithms using MapReduce. Our one-round lower bounds are proved in a very general model, which we call theMassively Parallel Communication (MPC)model, that allows processors to communicate arbitrary bits. Our multi-round lower bounds apply in a restricted version of the MPC model in which processors in subsequent rounds after the first communication round are only allowed to send tuples.
Source: Journal of the ACM
DOI: 10.1145/3125644
Authors: Paul Beame, Paraschos Koutris, Dan Suciu
Institution: University of Washington
Language: English
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
Open Access status: bronze
Awards: AF: Small: Communication and Resource Tradeoffs, AF: Small:Tradeoffs among Measures in Computational and Proof Complexity, III: Small: Query Compilation on Probabilistic Databases, III: Small: BeliefDB - Adding Belief Annotations to Databases, AitF: FULL: Query Processing with Optimal Communication Cost, BIGDATA: Mid-Scale: DCM: A Formal Foundation for Big Data Management
OpenAlex ID: W2765269579
