Hide menu

A Performance Optimization for Query-Based Data Access Interfaces on the Web

This project is designed for master students.

Description: Billions of statement-like data items exist in thousands of knowledge graphs on the Web, but few of those graphs can be queried _live_ from Web applications. That is, only a limited number of knowledge graphs are available in a queryable interface, and existing interfaces can be expensive to host at high availability. To mitigate this shortage of live queryable Web data, in earlier work we have specified a low-cost interface for servers, and a client-side algorithm that evaluates SPARQL queries against this interface (where SPARQL is a query language for knowledge graph data that is similar to SQL for relational data). The goal of this project is to design, implement, and evaluate a performance-enhancing extension of our interface and client-side query algorithm. More specifically, the extension will be based on the idea of Bloom filters and it will allow the client to retrieve fewer data from the server during the process of executing a given SPARQL query. Technologies involved in this project are a) RDF and SPARQL, as the data model and the query language for knowledge graph data (the basics of these should not be too difficult to learn), b) Javascript for the client-side algorithm, c) Java for the server-side implementation, and d) the concept of Bloom filters for the extension.

Prerequisites: Students who want to work on this project should have read a course on Database Technology. Additionally, familiarity with Java and with JavaScript is another requirement.

Contact: Olaf Hartig.

Page responsible: Olaf Hartig
Last updated: 2018-05-18