Serverless Computing Model Using Kubernetes and Knative in a Scalable Cloud Development

With the advancement of technology, more organizations are transitioning to Cloud Computing to meet their infrastructure needs. The popularity of this technology can be attributed to its numerous advantages, including cost-effectiveness and scalability. One focus of Cloud Computing is Serverless Computing, which provides a usage-based system allowing users to allocate costs only for the resources used, thereby enhancing cost efficiency and resource utilization. The open-source Knative Framework project was developed to adopt Serverless computing, adding components to Kubernetes to build a Serverless system. Using the Knative Framework, users can implement serverless services without relying on other cloud providers. This research aims to implement a Serverless system using Knative and conduct performance testing with parameters such as response time, resource utilization, and failure rates. The research results in a functioning Serverless system alongside Kubernetes and a usable monitoring system. It is found that the resource utilization performance of serverless CPU and RAM is similar to that of conventional systems. Additionally, the Cold Start time for the Serverless system is found to be 2.28 seconds. The Serverless system successfully performs automatic scaling based on incoming traffic demand.