Exam essentials

Autoscaling

Autoscaling is the ability for applications and infrastructure to automatically grow and shrink to meet demand. When demand goes up, more application instances and more infrastructure can be dynamically added to the environment. When demand drops, application instances and infrastructure can be removed. Kubernetes has a horizontal pod autoscaler to add and remove application instances to meet demand. It also has a cluster autoscaler that integrates with cloud platforms to add and remove nodes to meet infrastructure requirements. Metrics are used to make scaling decisions. Basic metrics include CPU and memory, whereas custom application metrics include things such as queue length and request response time.

Serverless

Serverless is a modern event-driven computing model where small functions are associated with specific events and only execute when those events occur. Sometimes we call it Function as a Service (FaaS) and many public clouds offer proprietary...