AWS provides many services which enable you to build whole topographies, including machines of all sizes (from t1.micro
to h1.4xlarge
), load balancers, file hosting, databeses, and CDNs, among other services.
The most correct configuration to you is very dependant on the expected traffic, budget, complexity of code, whether it is CPU heavy, memory heavy, or network heavy, and a lot of other specifics.
I suggest you start with a single instance of m3.medium
(a general purpose machine), do your benchmarks, so you can decide if you need to scale up or scale out.