Technical Report CSI-0022

Big Data Analytics using Graph Algorithms on Graph Databases

P. Balaghan and K. A. Hawick

Archived: 2016


Graph theory provides an important set of algorithms and tools for applying data analytics to big and complex data sets. Databases that make use of fundamental models that are not restricted to the traditional SQL relational model are becoming more prevalent. These so-called No-SQL databases come in a variety of forms but a particularly important class are the so-called graph databases which can model relationships as edges between information nodes in an graph-like structure. As well as offering a different way to model relationships that are not constrained by table structures, and various approaches to distributed data federation, graph databases also offer powerful opportunities to utilise many well-known graph computational algorithms to analyse patterns and properties of datasets. In this article we explore how well existing graph database packages support some of the better known graph algorithms, and how they can be implemented in a system such as Neo4J. We report on implementation issues as well as giving some performance and scalability results. We show how some of the accompanying graph apparatus that come with the software package can be used as well as speculating about implementing innovative graph algorithms not traditionally used in the context of analysing big data sets.

Keywords: graphs; graph algorithms; graph databases; interfaces; scalability; performance

Full Document Text: Not yet available.

Citation Information: BiBTeX database for CSI Notes.

BiBTeX reference:

        Title = {Big Data Analytics using Graph Algorithms on Graph Databases},
        Author = {P. Balaghan and K. A. Hawick},
        Institution = {Computer Science, University of Hull},
        Year = {2016},
        Address = {Cottingham Road, Hull HU6 7RX, UK},
        Month = {June},
        Number = {CSI-0022},
        Type = {CSI},
        Keywords = {graphs; graph algorithms; graph databases; interfaces; scalability; performance},
        Owner = {kahawick},
        Timestamp = {2016.11.07}