0% completed
Curious why there's a functional requirement that "the rate limit should be cons...
Z Wang
Mar 20, 2022
Curious why there's a functional requirement that "the rate limit should be considered across different servers." So we're assuming that the rate limiter lives close to application servers. Why not have a rate limiter in a reverse proxy layer that then directs to application servers? Then, we wouldn't need to have application servers concurrently access Redis, but perhaps multiple reverse proxy servers may need to instead? Or because application logic like user and API shouldn't be used at the reverse proxy layer?
0
0
Comments
Design Gurus4 years ago
The functional requirement should be considered independent of the actual design.
What the requirement really means is that if a client accesses same API on different servers, the rate limiting should be applied on that API access (across all servers) and not per serve...