Would anyone care to put forward a solution to proactively tackle a similar script.

What I am specifically after, methods to know that this one computer ( keep it simple ) has sent x 1,000 requests in a short time ( ie to quick to be human ).

Before some of you lay the blame purely on AT&T for having poor code.

Other scenarios which are similar but different.

Perhaps we want to use this to throttle api requests, or to tackle a brute force attempt on the login.