Some considerations – how do users interact with a variable number of servers?
After the previous explanations, Harold is thinking about a more philosophical question: how to use the existing applications, which were never designed for the cloud, in the new auto scaling environment. This question will lead to lots of interesting possibilities and many improvements in high availability and security.
Harold: Thinking about EC2 auto scaling, I have an important doubt.
Charles: Shoot.
Harold: Let’s imagine a real but simple scenario. We have our company web server, with all those web pages served using HTTPS from EC2 instances. Remember, these are dynamic pages, with our price list, products, catalog, orders, and so on. So, we can’t use S3 for them.
Alex: Yes, I remember we commented to create a web server in EC2, similar to the one we had on-premises.
