Throttling

is a technique that limits the number of requests or data that a server can handle in a given time period.

Helps to prevent overloading, improve performance, and allow quota limits.
Protects against denial-of-service (DoS), by limiting the number of requests per IP.
Helps to balance between different servers or regions.
Can be used to establish the priority of requests (pay for quality of service).
Can be used to govern and commercialize APIs (ex: Google Maps).
Cloud services use throttling for limits and quotas for storage, computing, and so on.