Java, Clouds and Cost Driven Architectures
In on-premise deployments, the networking costs are negligible, and a project is usually paying a fix per CPU / per RAM or even per server fee. You can entirely focus on the implementation of business logic and the fulfillment of functional requirements.
In public clouds, you will have to pay for the virtual hardware and the storage, bandwidth, API calls, IP addresses, and even network equipment. An additional non-functional requirement: the monthly costs becomes an essential non-functional requirement.
In this keynote, I would like to discuss various Java-based architectures from monoliths over microservices to lambdas deployed as virtual machines, containers, serverless environments, or even functions.
I will use various AWS services, from EC2, over ECS, Fargate, and EKS to AWS Lambdas for the Java deployments to give you a rough estimate. Your questions are highly appreciated.