Evaluation of Scalability and Multi-tenancy: A Use-Case

Flexibility and affordability in modern computing systems can be achieved through scalability and multi-tenancy of all involved devices. This paper presents a use-case of a multi-tenant software solution, where each tenant generates a stream of data that is subsequently processed. The processing output contains metrics to be presented on an analytical dashboard. The tenants are web applications developed using Ruby on Rails, the data streams ingestion and processing platform are developed using a Kafka cluster and KSQL, the analytical dashboard is developed using NodeJS. The use-case is evaluated by measuring the scalability through the response time from data stream creation to metrics display. The experiments are specified by changing the number of tenants and the workload for each tenant. The results show that the system can scale with regard to the number of tenants. The response time increases by 10% to 20%, when the number of tenants increases by 100% to 150%. With regard to the number of data items within the data stream, the response time is stable.