A GPU implementation of the Dynamic Yannakakis Algorithm

MPhil Thesis Defence


Title: "A GPU implementation of the Dynamic Yannakakis Algorithm"

By

Mr. Dexter LEE


Abstract

With the need to perform data updates on database systems that are increasing 
in size, researchers are looking for ways to evaluate join queries efficiently 
under frequent changes to the database. Recently, Idris et al. presented the 
Dynamic Yannakakis algorithm as one method to solve this problem. In this 
thesis, a GPU-based implementation of the algorithm is presented, which uses 
parallelism to further speed up the process. The experimental results comparing 
the performance of the sequential version and the parallel version of the 
algorithm are also provided. It is shown that when the data input size is large 
enough, the parallel version will achieve a better running time than the 
sequential version.


Date:			Friday, 24 August 2018

Time:			10:00am - 12:00noon

Venue:			Room 3494
 			Lifts 25/26

Committee Members:	Dr. Ke Yi (Supervisor)
 			Dr. Qiong Luo (Chairperson)
 			Dr. Sunil Arya


**** ALL are Welcome ****