Implementing the T-Systems KPI Dashboard

OData support
Benedek Zoltán
Department of Automation and Applied Informatics

Data visualization is one of the most difficult part of the IT world, especially during system supervising and monitoring. The scope of displayed information is extremely multilayered, because it depends on where you want to use (different devices, and platforms).

It is required to display different data on unified layout, which is easy to use and perfect for as a manager information system, and not at least it looks fine.

T-Systems Hungary Ltd. competence centre have started a project to satisfy these requirements four years ago. They create a web application, named KPI Dashboard with several iteration. The layout could be devided into three main categories. The first one is customized dashboard, second is categorized KPIs page, and the last one is the settings pages. This scenario expanded with new requirements, but the original project could not meet these requirements. These requirements are the followings: universal layout to different devices (mobile, tablet, desktop), new and more complex user and KPI policy management, new diagram visualization framework creator. The new framework required to rewrite source code.

Purpose of my thesis is to develop a new KPI Dashboard and take account of accepted principles, deficiencies and expected modern web application functionality.

This new application’s architecture is based on three-layer (Data Access Layer, Business Logic Layer, View), because the data come from different databases, for example. The control layer is responible for generalization of view. It is based on event system. Model layer communicates with the server’s BLL via AJAX. BLL is reachable on REST API interface. Response returns in JSON format if user meets the policy requirement. The new application will be clear, owing to restriction of RESTful and modular structure.


Please sign in to download the files of this thesis.