FinOps for cloud provider ?

[agonzalez]

Hello,

Is anyone doing cloud cost optimization / FinOps on GCP (or other cloud providers) ?

One of my customer runs on GCP wants to right-size their Kubernetes workloads (and other GCP services...). The ideal output would look like:

• Workload X

Current: 

• CPU/RAM:
  • Request
  • Limits
  • Avg est cost: x$/month

• Optimized:
  • CPU/RAM:
    • Request
    • Limits
    • Avg est cost: y$/month
• Result: Downsize Memory 8Gb --> 4Gb
• Estimation cost saving: z$/month

Has anyone built something like this ? dashboard, notebook, workflow ?
And if yes, how did you do it ? 

Thanks for your help

[sujit_k_singh]

Hi @agonzalez 

Dynatrace actually has built-in capabilities for this exact use case:Dynatrace provides workload resource optimization out of the box. Go to Kubernetes → Workloads — it shows CPU/Memory requests, limits, and actual usage per workload with right-sizing recommendations.
Optimize workload resource usage with Kubernetes app and Notebooks — Dynatrace Docs

For the exact output format you described, use Notebooks with DQL . like:

fetch dt.entity.cloud_application
| fieldsAdd requests_cpu = ..., limits_cpu = ..., actual_cpu = ...
| fieldsAdd recommended_cpu = ...

Cost & Carbon Optimization App

Dynatrace has a dedicated Cost & Carbon Optimization app that calculates IT cost based on public cloud price lists (including GCP) at host level.

Cost & Carbon Optimization — Dynatrace Docs

Thanks,

Sujit