A Graphical Compute Infrastructure Modelling Tool

With the rise of DevOps and Cloud Computing, the interest in automating everything around the development and operation of software grew. Nowadays, continuous deployment and automated infrastructure provisioning are more and more common practices. A less considered aspect are the costs generated by the underlying infrastructure. Historically, the procurement of IT infrastructure was an elaborate process and required upfront investments that could be substantial. With cloud providers, no upfront investments are required, and infrastructure is available within a short time after a click or API call. This offers a great flexibility for scaling software services and allowing developers to quickly get infrastructure for the development. However, managing the costs has become a much more complex task.

A research project at the Research Group Software Construction explores approaches for forecasting future cloud computing infrastructure costs based on infrastructure need forecast provided, for example, by people with in-depth knowledge of the respective software services and how they are expected to develop in the near future.

In this thesis project, a tool should be developed that allows its users to graphically model compute infrastructure needs, e.g., needs for virtual machines and block storage, based on an existing textual specification language.

The focus and scope will be defined depending on whether the thesis is a bachelor or master thesis. The research questions will be developed in collaboration with the supervisor in the introduction period before the thesis registration.

It is recommended to have a basic understanding of what services cloud providers offer and how they charge for their services. This knowledge can be obtained by familiarizing with one or two popular cloud providers, e.g., Amazon Web Services (AWS) or Google Cloud Platform (GCP).

Knowledge and experience in

  • web development (HTML, CSS, JavaScript) and web technologies (e.g., Angular), and
  • docker

is recommended.

If you are interested in this topic, but are unsure whether you have the knowledge and skill for this thesis project, contact me anyway and let me know.

Project information

Status:

In progress

Thesis for degree:

Bachelor

Student:

Savas Köklü

Supervisor:
Id:

2023-011