Ok lets think about that — how many concurrent request can accommodate our design architecture ? Can we say 1 server accommodate max 10K request ?

So if we check our table you can see that;

So according to our e-commerce application, we can elaborate these non-functional requirements;

Scalability = e-commerce application should able to serve millions of users

Ok, now lets talk about acceptable latency, how we can make our application for acceptable latency, if our application get use more and more users ?

As you can see in the table, we will start a small e-commerce application that get only 2K concurrent user and gets 500 request per second. But how we can scale the application if we need to handle more users in our application ?