Concept and implementation of a pluggable framework for storage, transformation, and analysis of large-scale enterprise topology graphs

Abstract

The addition of on-demand cloud computing offerings increases the complexity of IT systems rapidly. Enterprise Topology Graphs depict all the components of enterprise IT and their relations, to regain insight into enterprise IT. The focus of this work is the research, design and implementation of a framework to store and manage these graphs in an efficient way. The difficulties are the enormous graph sizes and a lot of meta information, leading to a complex design to offer a good performing solution. The framework is based on a graph database to store the Enterprise Topology Graph efficiently and offers a pluggable architecture to be able to extend the functionality, e.g. with transformation operations the graphs. With the reference implementation of this framework, the complex structures of enterprise IT systems can be stored, managed and easily manipulated to gain a more detailed view on the IT components and their dependencies.