gRPC HVA scaling.plantuml

' SPDX-License-Identifier: Apache-2.0

@startuml gRPC+HPA scaling
queue "Service 1" as s1
control "Linkerd Proxy" as lp
database "LinkerD Control\nPane" as cp
queue "Service 2" as s2
control Kubernetes as k8s

s1->lp: gRPC request
note over lp
Linkerd will know all 
Pods that's been created
with the same deployment 
file. It will then do 
Load Balancing to all 
the Service 2 instances 
running.
end note
lp->cp: Telementry + \ncontrol signals
lp->s2: gRPC request
k8s-->k8s: monitor s1 and \ns2 resources
note over k8s
If CPU reaches x% or 
RAM use reaches x%, 
k8s will use the 
deployment file to 
deploy another pod.
end note
@enduml