Hey Abhishek, appreciate the reply and insights. And yes, I agree, that was how I found it difficult to pick between A and D. I was considering how much compute resources must be taken up for them to outgrow their on-premise capacity, and then I was looking at the “most scalable and most cost effective” requirement. I was torn then, as for scalability at that much legacy compute and the need to have the capacity on-demand (multi-player game, so what wiggle room is there for accommodating any warm-up time to scale out more EC2) so I erred towards Lambda. Then it was the balance of how much could this company spot and reserve to keep the costs of EC2 down and still be instantaneously scalable for the gamer demand (versus Lambda). That’s why I ended with D, but I was definitely torn between the two, and I’m thanking you for explaining the rationale from someone picking A — it challenges my thought process. Thanks Abhishek.