Abstract: Software Elasticity is the concept of adapting available resources to the current or expected workload. This concept fits modern and stateless microservice architectures, which are scalable by design. Their scalability is closely related to Software Resilience and places new demands on cloud architectures. The JValue Open Data Service (JValue ODS) is an open data platform with focus on Extract, Transform, Load (ETL) pipelines and aims to make the usage of open data easy, reliable and safe. For the success of the ODS, an Elastic and therefore Resilient hosting is mandatory. This thesis deploys the ODS to an on-premise Kubernetes cluster to improve the uptime guarantee, discusses different deployment strategies, elaborates horizontal microservice scaling techniques and operates the necessary infrastructure. This thesis presents Peffer’s Design Research Process to build a concept for Elasticity in microservice-based architectures. The concept is demonstrated and evaluated in the context of the JValue ODS.
Keywords: Microservices, elasticity, scalability, kubernetes, devops
PDF: Master Thesis
Reference: Aron Metzig. Elasticity Concept for a Microservice-based System. Master Thesis. Friedrich-Alexander-Universität Erlangen-Nürnberg: 2022.