Hello
I have been noticing performance inconsistencies in my Kubernetes cluster when certain deployments scale unevenly. 
For eg; when some pods get scheduled on nodes with more available resources while others are placed on more constrained nodes; the applicationâs response time varies noticeably. 
Even though auto-scaling is working; the distribution of load & resources doesnât feel balanced.
I am wondering if this is something that can be mitigated with better resource requests/limits; pod affinity rules / even smarter HPA/VPA configurations. 
Have others faced similar issues where uneven scaling impacted real-world app performance? If so; what strategies / tuning options worked best for you?
I have gone through the Kubernetes Scheduling & Eviction guide but would appreciate practical insights from this community, especially around making scaling more predictable and fair across nodes. While learning more about distributed systems; I also explored what is dax in power bi to understand how different platforms handle calculations & performance at scale.
Thank you !!