[[Serverless computing]] has a few key benefits:
No server management You don't have to handle the infrastructure, the cloud provider does. Less setup, less maintenance.
Lower costs Pay only for what you use. If you don't receive much traffic, you aren't paying to keep an idle server up
Easily scalable Serverless is inherently scalable, due to the way it's written. If the function needs to run more, it will. If it needs to run less, no problem.
Code can run closer to the user There's no base server, it's just functions that need to run somewhere. [[20221107042513-edge-computing]]
However there's also some drawbacks:
Debugging is harder You don't have a server - it's harder to replicate that in a local environment
Long-running processes Functions can only run for a certain time. If you need to process a lot of things, the cost saving angle becomes less appealing, and the time limit may become limiting.
Performance Serverless can suffer from 'cold starts', where it needs to boot up when there's been a long delay since last running.
[[architecture]] [[infrastructure]] [[microservices]] [[serverless]]