The biggest thing to keep in mind when it comes to the scalability of Azure OpenAI service are the quota and limits. I'd start by taking a look at this page that talks about the limits. It's also important to learn how to manage the quota for your service.

In terms of performance, the best thing I can recommend is trying out your prompts in Azure OpenAI to understand the performance. One big factor for performance is the length of your prompts, particularly the number of output tokens.