What was the query rate in systems with high load?

The metric of requests per second (RPS) in high-load systems heavily depends on architecture, hardware, and optimization. In real high-load projects, RPS can reach tens or hundreds of thousands.

For example:

• Web servers of large services can handle from 10,000 to 100,000 requests per second.
• In distributed systems with horizontal scaling, RPS can be even higher.

It is important to understand that achieving such metrics involves caching, load balancing, asynchronous processing, and code optimization.

Specific figures depend on the context and project requirements.