Master Thesis Proposal - Predictive Cloud-based Scaling Algorithms

In today's fast-paced business landscape, adaptability is the key to success. Are you tired of struggling to keep up with fluctuating workloads, leading to system crashes, downtime, and potentially frustrated customers? It's time to pave the way for the future of efficiency and help us develop the scalable systems of tomorrow!

We would like to eliminate the headaches of managing server capacity and ensure optimal performance, even during peak demand. We want to develop a system that can automatically scale resources up or down based on real-time load and predict the stress of the system based on earlier data, ensuring applications run smoothly regardless of traffic spikes.

The benefit of this approach is to eliminate system crashes, downtime and overspending on infrastructure. With a scalable system you only need to pay for the resources you use, reducing operational costs while enhancing performance.

For this master thesis we would like to implement and evaluate different scaling algorithms that are relevant based on current studies and prior master theses. The goal of the algorithms is to predict the oncoming load of a system based on collected data and scale the system accordingly. Additionally, we foresee that different load patterns and algorithms can be tested and compared, based on, for example, machine learning, analysis in the frequency domain, etc. Finally, the implemented algorithms are evaluated and compared to industry-standard scaling algorithms in a cloud environment.

Don't let fluctuations in workload hold your back. Embrace the power of automatic scalability and ensure seamless performance, no matter the demands. Together, we'll drive efficiency and success to new heights, join us now!

Proposed research question:

• What is the efficiency of predictive scaling algorithms in a cloud environment?

Features:

• Scalable applications: Kubernetes clusters, containers
• Cloud based applications.
• Predictivity models
• Simulations

Background:

• We foresee that a student with a background in software engineering will excel in the proposed assignment.
• We will only accept one student per assignment.

Application:

We look forward to receiving your resume, and preferably, a personal letter in which you explain why you want to write your thesis with Syntronic.

We screen and evaluate applications on an ongoing basis.