Skip to main content

Resource Analytics

The resource analytics section provides intelligent analysis of pod distribution, namespace resource usage, and consumption patterns to optimize cluster performance and capacity planning.

Pod Distribution Analytics​

Pod Status Distribution​

![Figure needed]

Screenshot of pod distribution analytics section

Summary Statistics​

  • Total Running Pods: Count of active pods across the cluster
  • Most Loaded Node: Node with the highest pod count
  • Total Containers: Aggregate container count across all pods
  • Active Namespaces: Number of namespaces with running pods

Distribution Metrics​

  • Average Pods per Node: Calculated average distribution
  • Pod Range: Minimum and maximum pods per node
  • Distribution Balance: How evenly pods are spread across nodes
  • Utilization Efficiency: Pod allocation efficiency metrics

Resource Consumption Analysis​

Pod Resource Categories​

![Figure needed]

Screenshot showing resource consumption categorization

High CPU/Memory Pods:

  • Definition: Pods consuming significant CPU or memory resources
  • Threshold: Typically over 1 CPU core or over 2GB memory
  • Count Display: Number of high-resource pods
  • Impact: Pods that significantly affect cluster capacity

Medium CPU/Memory Pods:

  • Definition: Pods with moderate resource consumption
  • Range: 0.5-1 CPU core or 1-2GB memory
  • Workload Type: Standard application workloads
  • Balance: Target category for most workloads

Low CPU/Memory Pods:

  • Definition: Lightweight pods with minimal resource usage
  • Range: under 0.5 CPU core or under 1GB memory
  • Examples: Monitoring agents, logging pods, sidecars
  • Efficiency: Good for resource optimization

Resource Metrics​

  • Average Pod Resources: CPU and memory per pod averages
  • Resource Distribution: Statistical analysis of resource spread
  • Consumption Patterns: Identification of usage patterns
  • Optimization Opportunities: Areas for improvement

Namespace Resource Distribution​

Top Namespace Analytics​

![Figure needed]

Screenshot of namespace resource breakdown showing top 6 namespaces

Information Displayed​

The dashboard shows the top 6 namespaces by resource consumption:

Per Namespace Metrics:

  • Namespace Name: Kubernetes namespace identifier
  • Pod Count: Number of active pods in the namespace
  • CPU Usage: Aggregated CPU consumption for the namespace
  • Memory Usage: Aggregated memory consumption for the namespace
  • Resource Percentage: Relative usage compared to cluster total

Namespace Categories​

  • System Namespaces: kube-system, kube-public, etc.
  • Application Namespaces: User application deployments
  • Monitoring Namespaces: Observability and monitoring tools
  • Infrastructure Namespaces: Supporting infrastructure services

Distribution Insights​

Resource Balance Analysis​

  • Concentration: How resources are concentrated across namespaces
  • Distribution Equity: Whether resources are evenly distributed
  • Hotspot Identification: Namespaces consuming disproportionate resources
  • Capacity Impact: Which namespaces drive capacity requirements

Planning Information​

  • Growth Patterns: How namespace resource usage grows over time
  • Scaling Indicators: Namespaces that may need resource scaling
  • Optimization Targets: Namespaces with optimization potential
  • Cost Allocation: Resource usage for cost attribution

Analytics Interpretation​

Pod Distribution Analysis​

Optimal Distribution Patterns​

Good Distribution:
- Even spread across nodes (±20% variance)
- No single node over 80% pod capacity
- Balanced resource utilization

Problematic Patterns:
- High concentration on few nodes
- Some nodes underutilized while others overloaded
- Frequent pod evictions due to resource pressure

Load Balancing Assessment​

  • Node Utilization Variance: Difference between highest and lowest loaded nodes
  • Pod Placement Efficiency: How well pods are distributed
  • Resource Hotspots: Nodes with disproportionate resource usage
  • Rebalancing Opportunities: Potential for better distribution

Resource Consumption Patterns​

Healthy Consumption Profile​

Typical Healthy Cluster:
- 60-70% medium resource pods
- 20-30% low resource pods  
- 10-20% high resource pods
- Balanced distribution across nodes

Optimization Indicators​

  • High Resource Concentration: Too many high-resource pods
  • Underutilization: Too many low-resource pods with excess capacity
  • Imbalance: Uneven resource distribution across nodes
  • Waste Indicators: Allocated but unused resources

Use Cases​

Capacity Planning​

Growth Projection​

  1. Trend Analysis: Analyze resource consumption trends
  2. Namespace Growth: Project growth by namespace
  3. Resource Requirements: Calculate future capacity needs
  4. Scaling Timeline: Plan when to add capacity

Resource Allocation​

  1. Namespace Quotas: Set appropriate resource quotas
  2. Node Sizing: Determine optimal node configurations
  3. Cluster Scaling: Plan horizontal vs. vertical scaling
  4. Cost Optimization: Optimize resource allocation for cost

Performance Optimization​

Workload Distribution​

  1. Anti-affinity Rules: Improve pod distribution with scheduling rules
  2. Node Selection: Guide pod placement decisions
  3. Resource Requests: Optimize pod resource requests and limits
  4. Load Balancing: Ensure even resource utilization

Resource Right-sizing​

  1. Over-provisioning: Identify over-allocated resources
  2. Under-provisioning: Find under-allocated workloads
  3. Optimal Sizing: Right-size pod resource specifications
  4. Efficiency Gains: Improve overall cluster efficiency

Troubleshooting​

Performance Issues​

  1. Hotspot Analysis: Identify resource hotspots
  2. Bottleneck Detection: Find resource bottlenecks
  3. Imbalance Resolution: Address uneven resource distribution
  4. Capacity Problems: Diagnose capacity-related issues

Resource Conflicts​

  1. Namespace Conflicts: Identify competing namespaces
  2. Resource Pressure: Find sources of resource pressure
  3. Scheduling Issues: Understand pod scheduling problems
  4. Performance Degradation: Analyze performance impacts

Analytics Best Practices​

Regular Analysis​

  1. Weekly Reviews: Analyze resource analytics weekly
  2. Trend Monitoring: Track changes over time
  3. Pattern Recognition: Identify recurring patterns
  4. Anomaly Detection: Spot unusual resource behavior

Optimization Actions​

  1. Resource Requests: Adjust based on actual usage
  2. Pod Placement: Improve distribution through affinity rules
  3. Namespace Limits: Set appropriate resource limits
  4. Scaling Decisions: Make informed scaling choices

Proactive Management​

  1. Capacity Alerts: Set up alerts for resource thresholds
  2. Trend Alerts: Monitor for unusual trend changes
  3. Efficiency Metrics: Track resource efficiency over time
  4. Cost Monitoring: Monitor resource costs and optimization

Data Sources and Accuracy​

Metrics Collection​

  • Source: Kubernetes metrics server and API
  • Frequency: Real-time collection with 30-second aggregation
  • Accuracy: High accuracy for capacity planning decisions
  • Coverage: Complete cluster resource visibility

Data Processing​

  • Aggregation: Intelligent aggregation across nodes and namespaces
  • Calculation: Real-time statistical calculations
  • Filtering: Automatic filtering of system vs. user workloads
  • Validation: Data validation and consistency checking

Next: Learn about Resource Leaderboards for identifying top resource consumers.