What are the advantages of traditional servers compared to serverless architecture? For example, does serverless make sense for applications that need to run continuously? In projects with critical performance requirements, in what situations should it be preferred?
Who should prefer serverless architecture?
👁️ 7 views💬 2 replies❤️ 0 likes
2 Replies
Serverless ever crossed your mind as a solution for continuously running applications? Like a 24/7 accessible dashboard or API—how do you even get past the "cold start" problem in serverless?
I think serverless architecture is ideal for applications with unpredictable or sudden traffic spikes. For example, I set up the backend of a new mobile game using AWS Lambda last month. The game's users spike in traffic when they go to work at 8 AM and return at 6 PM. Since Lambda scales automatically, it can go from 100 users at midnight to 10,000 by 8 AM, and since you only pay for usage, it saves millions of TL in costs.
However, serverless can be a bit risky for systems that require critical performance. For instance, if a banking service needs to respond in under 200ms, it’s more logical to opt for a constantly running container or physical server. I also moved the payment module of an e-commerce project to Kubernetes because the "cold start" delay in Lambda sometimes reached a second, causing issues with receiving payments from customers. Ultimately, every tool has its place; you just need to analyze your needs properly.