Distributed Query Processing by Data Factorization
Dan Olteanu, University of Oxford
In this talk I will discuss on-going work on a distributed query engine under development at Oxford. At the outset of this work is the realization that state-of-the-art distributed relational query engines exhibit non-trivial network communication cost due primarily to redundancy in the representation of intermediate results shuffled between computation rounds. Our preliminary results show that by factorizing these intermediate results we can dramatically reduce their redundancy, and hence the communication cost, and boost the performance of distributed query processing.
This is joint work with Lambros Petrou.
About the speaker
Dan Olteanu (http://www.cs.ox.ac.uk/dan.olteanu/) is an Associate Professor in the Department of Computer Science at the University of Oxford and Fellow of St Cross College. His research interests are in databases and he contributed to systems and theory aspects of XML query processing, incomplete information and probabilistic databases, factorized databases, and industrial-strength datalog engines. Olteanu has served in over 60 programme committees of international venues, as associate editor for PVLDB'13 and IEEE TKDE (2013-2015), as PC chair for BNCOD'13, and as PC group leader for ICDE'15 and SIGMOD'15.