Hi All,
I'm running a cluster, 160+ nodes, 700+ namespaces. 16c/32Gi nodes.
Using latest AG 1.25, Operator 1.7.x ...
we are just using for Routing, Kubernetes Monitoring.
Application Only (csi)+ Kubernetes - not fullstack
We're running OTEL separately for Prometheus scraping and tracing.
Looking to see if anyone might have a cluster that is equivalent or bigger?
If so, what type of resources are you provisioning for the active gate, or what solutions have you had to put in place to address high resource utilisation. Or have you had to go the old dedicated VM for the Active Gate in larger environments.
Our AG is currently sitting at a base line of 10 Core, 10 GI minimum for requests and topping out at 15 core and 20 Gi + mem. It just seems excessive and would like to validate against the community.
I don't really want to have to look at dedicated supersized nodes just to keep Dynatrace going or move back to a dedicated VM (we were previously 8c / 32Gi VM which had more than enough room to breathe).
The only other contender I've found so far for this type of resource utilisation is the Atlassian suite.
