Serverless architecture, refer to the application that significantly depends on third party services or custom code that’s run with in ephemeral containers, AWS Lambda is the best-known example.
As the name suggests it not like that code is running without servers, here the code written in traditional server style such as micro services so pricing the based on the actual amount of resources consumed by application rather than on pre-purchased units of capacity.
Cost, effective because it not like renting or purchasing fixed quality of servers even it’s more cost effective than provisioning an autoscaling group.
Operations, as developers and operators do not need to spend time for setting up and tuning auto scaling policies.
Productivity, the units of code exposed to the outside world are simple functions. means programmer does not need to worry about multithreading or HTTP directly requests.
Performance, Infrequently used serverless code may suffer from greater response latency than a dedicated server.
Monitoring and debugging, diagnosing performance or resource usage problem may be more difficult here, attaching profilers or debuggers little difficult. furthermore, the environment in which the code runs is typically not open source so its performance characteristics cannot be precisely replicated in the local environment.
AWS Lambda, Azure functions with node.js, Kubernetes etc.