Load Balancer's Buffering and Scheduling Strategies

# Describe the question
I am studying the internal operations of OpenWhisk, particularly the load balancer. I would like to understand more about the buffering and scheduling strategies for incoming requests.

# Details to consider

1. How does the buffering of incoming requests work within the load balancer? Is there a dedicated request buffer?
What are the factors taken into consideration in the load balancer's scheduling strategy when distributing these buffered requests to the invokers?
2. In situations where invoker load is high, how does the system respond? Are more requests buffered during these times?
3. In situations where invoker load is high, how does the system respond? Does the load balancer consider waiting for a period of time before issuing a request immediately when the invoker is always available?
# Additional context
Understanding these details will be beneficial for optimizing my use of OpenWhisk and contributing to the project. Any information or resources you could provide would be greatly appreciated.

Thank you!

The issue names no files, tests, or entry points. Begin by locating the OpenWhisk load balancer's buffering and scheduling implementation and existing documentation; done means providing a project-specific explanation of request buffering, invoker selection, and high-load behavior.

Written by the indexing model from the issue text.