I have a quick question on the LEAST_CONN load balancing method.
describes LEAST_CONN as
The least request load balancer uses an O(1) algorithm which selects two random healthy hosts and picks the host which has fewer active requests.
How does istio ingress controller know the number of active requests in a pod? Are these constantly programmed by “pilot” based on stats received by the “mixer” from the “istio proxy” in the pod ?
What happens if the ingress controller has scaled and has 5 pods? Does the mixer update all the ingress controllers? Doesn’t this become too performance intensive?
Or am I completely wrong? And the ingress controller knows about the number of active requests in a pod in a different manner?