Utilising modern hardware for skyline and skycube computation: From the CPU to the GPU and back again

PhD defence, Wednesday 1 February 2017, Kenneth Sejdenfaden Bøgh

2017.02.01 | Rita Høyer

Kenneth Sejdenfaden Bøgh

During his studies, Kenneth Sejdenfaden Bøgh investigated how modern hardware such as Graphics Processing Units (GPUs) and multicore Central Processing Units (CPUs) may be utilised to compute the skyline and skycube database operators efficiently. 

In the past decade, GPUs have evolved to become massively parallel, general purpose computing devices and CPUs have changed from executing tasks sequentially to doing so in parallel. Kenneth studied how these changes in hardware design might be utilised to accelerate two expensive database operators, called the skyline and the skycube, while retaining efficiency in the algorithmic design, and making algorithms portable between the GPU and the CPU.

The new research findings contribute to the understanding of how algorithms may be designed to run on either CPUs, GPUs or both, as well as how how skylines and skycubes may be computed on large data sets, while retaining good response times.

The PhD degree was completed at the Department of Computer Science, Science and Technology, Aarhus University.

This résumé was prepared by the PhD student.

Time: Wednesday 1 February 2017 at 12:30
Place: Building 5342 (Ada), room 333, Department of Computer Science, Åbogade 34, Aarhus University, 8200 Aarhus N.
Title of dissertation:  Parallel Algorithms for Skyline and Skycube Computation on Multicore CPUs and GPUs
Contact information: Kenneth Sejdenfaden Bøgh, e-mail: ksb@cs.au.dk, tel.: +45 31602515
Members of the assessment committee:
Associate Professor Katja Hose, Department of Computer Science, Aalborg University, Denmark
Professor, Dr Jens Teubner, Department of Computer Science, Technische Universität,  Dortmund, Germany Associate Professor Anders Møller, Department of Computer Science, Aarhus University (chair)]
Main supervisor:
Associate Professor Ira Assent, Computer Science, Aarhus University
Language: The PhD dissertation will be defended in English

The defence is public.
The dissertation is available for reading at the Graduate School of Science and Technology/GSST, Ny Munkegade 120, building 1520, rooms 128-134, 8000 Aarhus C.

PhD defence
Comments on content: 
Revised 25.09.2017