SPARQL Query Optimization with Pavel Klinov
Date: 28 Jan 2021
Time: 6pm Berlin, 5pm Dublin and 12pm New York and 9am San Francisco Times
Location: Online zoom
Chapters: Berlin, Oslo, Milano, Zürich, London, New York, DC, Maryland, Boston and San Francisco
In this session Pavel will describe the SPARQL evaluation semantics and some of the main algorithms used for processing SPARQL queries, such as joins. You will learn some common optimizations used by SPARQL query engines to make queries run faster. You will also learn how to detect and avoid the pitfalls which make queries less amenable to optimizations. Finally Pavel will demonstrate some general and Stardog-specific techniques for performance debugging of SPARQL queries as well as the techniques to coerce the query engine to execute the query in a particular way.
The query optimization techniques and evaluation algorithms to improve SPARQL query performance discussed in this session can be viewed as an extension to a series of presentations by Pavel on the topic:
Pavel Klinov is a software engineer leading the Query Engine team at Stardog. In that role he is primarily responsible for the evolution of Stardog’s SPARQL query execution engine, particularly, the development of novel query optimization techniques and evaluation algorithms. Some of them extend beyond pure SPARQL, for example, path queries introduced in 2018 or, more recently, SPARQL*.
Pavel has been with Stardog since 2011 (including a period of parallel employment at the University of Ulm between 2012 and 2015). Prior to that he obtained his PhD in the area of probabilistic logics at the University of Manchester and worked as a software developer at CERN during the early days of the LHC project.
Session-Level: Intermediate-Advanced Session-Type: Technology-Standards-Performance