Native Kubernetes Integration
Deploy the DeepCost operator directly in your Kubernetes cluster for automated cost optimization with custom resources and policy-driven management.
Native Operator
Kubernetes-native operator that runs in your cluster
Custom Resources
CRDs for declarative cost optimization policies
Automated Optimization
Continuous optimization without manual intervention
Safe Operations
Built-in safety checks and rollback capabilities
Quick Installation
Install Operator
kubectl apply -f https://deploy.deepcost.ai/operator.yamlDeploy the DeepCost operator to your cluster
Create Secret
kubectl create secret generic deepcost-config --from-literal=api-key=your-api-keyConfigure your DeepCost API credentials
Apply Policy
kubectl apply -f cost-policy.yamlCreate your first cost optimization policy
Monitor Results
kubectl get costpoliciesWatch automated optimization in action
Custom Resource Definitions
CostPolicy
Define cost optimization rules and constraints
apiVersion: cost.deepcost.ai/v1
kind: CostPolicy
metadata:
name: webapp-optimization
spec:
selector:
matchLabels:
app: webapp
optimization:
rightsizing: enabled
spotInstances: preferred
autoscaling:
minReplicas: 2
maxReplicas: 20
targetCPU: 70
budget:
monthly: "$5000"
alerts:
- threshold: 80%
action: notify
- threshold: 95%
action: scale-downCostAlert
Configure budget alerts and notifications
apiVersion: cost.deepcost.ai/v1
kind: CostAlert
metadata:
name: budget-alert
spec:
budget: "$10000"
period: monthly
thresholds:
- percentage: 50
severity: info
- percentage: 80
severity: warning
- percentage: 95
severity: critical
notifications:
slack: "#cost-alerts"
email: ["[email protected]"]SpotPolicy
Manage spot instance usage and failover
apiVersion: cost.deepcost.ai/v1
kind: SpotPolicy
metadata:
name: spot-optimization
spec:
nodeGroups:
- name: workers
spotPercentage: 80
onDemandBase: 2
fallbackStrategy: immediate
workloads:
- selector:
app: batch-jobs
spotTolerance: high
- selector:
app: web-frontend
spotTolerance: mediumOptimization Capabilities
Pod Resource Optimization
Automatically optimize CPU and memory requests/limits
Horizontal Pod Autoscaling
Intelligent HPA configuration based on multiple metrics
Node Pool Management
Optimize node pool configuration and instance types
Spot Instance Integration
Intelligent spot instance usage with automatic failover
Advanced Kubernetes Features
Policy as Code
Define cost optimization policies using YAML and manage them through GitOps workflows.
Cluster Autoscaling
Intelligent cluster autoscaling that considers both cost and performance requirements.
Safe Operations
Built-in safety mechanisms, rollback capabilities, and performance monitoring.
Frequently Asked Questions
How long does it take to deploy the DeepCost operator?
Deployment takes about 5 minutes. Simply run kubectl apply with our operator manifest and create your first CostPolicy custom resource. The operator will immediately start analyzing your cluster.
What permissions does the DeepCost operator need?
The operator needs read access to pods, deployments, and resource metrics, plus write access to implement autoscaling and optimization changes. All permissions are scoped with RBAC following Kubernetes security best practices.
Is my Kubernetes cluster data secure?
Yes, the DeepCost operator runs entirely within your cluster. All cost data and optimization policies stay in your environment. Communication with DeepCost cloud services uses encrypted connections and follows SOC 2 compliance.
Can I use DeepCost with multiple Kubernetes clusters?
Absolutely. Deploy the operator in each cluster and manage them all from a unified dashboard. You get cross-cluster cost visibility, centralized policy management, and consolidated reporting across all your Kubernetes environments.
How quickly will I see cost savings?
Most customers see optimization recommendations within hours of deployment. With automated pod rightsizing, node pool optimization, and spot instance integration, typical savings of 40-60% are achieved within the first month.