Modern scalable distributed cloud native applications

Bc. Ondřej Svojše

Master's thesis

Abstract:

In the thesis there is examined the state of modern public cloud platforms and the services offered that enable them to operate highly scalable applications with lower costs. First part is focused on technology and cloud services. The way they shape current cloud native application development and the platform they provide for developers to build reliable and resilient solutions. Looking further there is examined elements of how technologies such as containers or orchestrators affect application architecture, exploring the benefits and challenges introduced with this new way of doing things. In the following part there is quantitative analysis of a developers survey conducted in a large Czech software company on topic. The aim of it is to find out the views developers have on learning new technologies, public cloud and key factors and requirements they view for delivering good SaaS product at scale. Next step is exploring the platform of a large cloud infrastructure provider. Based on ideas and patterns there is documented building of Kubernetes cluster with many services that are communicating with each other through asynchronous messages bus like Apache Kafka or directly through HTTP protocol and are horizontally scaling instances based on workload. Then there is shown the importance of CI / CD pipeline in delivery to achieve zero downtime deployments of individual services. Another part mentioned in the first part of research is the observability of platforms, the following part is examining ways how to trace the request going through the cluster and important metrics for keeping eyes on day to day operation. In the last part load test is performed and observation of performance are done. In the end we evaluate the objectives of the thesis and conclude the results.

Keywords

Public Cloud, Containerization, Orchestrators, Virtualization, Distributed Systems, Cloud Native Applications, Microservices, Monitoring, Tracing

Language used: English

Date on which the thesis was submitted / produced: 28. 3. 2021

Thesis defence

• Supervisor: doc. Ing. Vojtěch Merunka, Ph.D.

Czech University of Life Sciences Prague

Faculty of Economics and Management

Master programme / field:
Systems Engineering and Informatics / Informatics