Tuesday, Sept 24, 1 - 2 pm at the PITLab:
Panagiotis (Panos) Bouros
Interval Joins: Evaluation - Parallelization - Aggregation
The interval join is a basic operation that finds application in temporal, spatial, and uncertain databases. Given a 1D discrete or continuous space, an interval i = [start, end] is defined by a start and an end point in this space. The interval join of inputs R, S identifies all pairs of intervals r in R, s in S that intersect, i.e., r.start <= s.start <= r.end or s.start <= r.start <= s.end.
This talk will present my recent work on interval joins towards three directions. First, I target the in-memory evaluation of interval joins. Although a number of single-threaded methods have been proposed for this purpose, classic plane sweep approaches have not been considered to their full potential. Then, I shift my focus to the parallel execution of the join that benefits from the existence of multiple CPU cores in a system. I discuss solutions that either physically partition input data or force running threads to read from shared memory in parallel. Finally, I discuss aggregation of interval data; a new count semi-join operation is presented for selecting or ranking intervals based on the number of join pairs they appear in.
Panagiotis Bouros received his diploma and PhD degree from the School of Electrical and Computer Engineering at the National Technical University of Athens, Greece. Since 2018, he is an assistant professor at the Institute of Computer Science in Johannes Gutenberg University Mainz, Germany and the head of the Data Management group. Prior to Mainz, he held research positions at Aarhus University, Denmark, Humboldt-Universitaet zu Berlin, Germany and the University of Hong Kong, Hong Kong SAR, China. His research focuses on managing and querying complex data types including spatial, temporal and text, and on routing optimization problems.