Skip to main content

Metrics Dashboard

The metrics dashboard provides real-time cluster monitoring with comprehensive resource analytics and automatic data refresh capabilities.

Accessing the Dashboard​

  1. Go to the cluster details page
  2. Click on the "Metrics" tab in the cluster navigation
  3. The dashboard loads with real-time data and begins auto-refreshing

![Figure needed]

Screenshot of cluster navigation with Metrics tab highlighted

Dashboard Loading​

  • Initial Load: Dashboard loads current metrics data
  • Auto-refresh Start: 30-second refresh timer begins automatically
  • Connection Status: Visual indicators show data connection status

Auto-refresh Functionality​

Automatic Updates​

  • Refresh Interval: Data updates every 30 seconds automatically
  • Background Process: Updates occur without interrupting viewing
  • Visual Indicators: Subtle indicators during data refresh
  • Seamless Experience: No page reload or interface disruption

![Figure needed]

Screenshot showing auto-refresh indicator and last updated timestamp

Manual Refresh Controls​

  • Refresh Button: Click to force immediate data update
  • Last Updated: Timestamp shows when data was last refreshed
  • Status Indicators: Visual cues for data freshness and connectivity

Connection Management​

  • Error Handling: Graceful handling of temporary connection issues
  • Retry Logic: Automatic retry for failed refresh attempts
  • Status Display: Clear indication of connection problems

Cluster Resource Overview​

Summary Cards​

The dashboard begins with high-level cluster statistics:

![Figure needed]

Screenshot of summary cards showing total nodes and running pods

Total Nodes Card​

  • Node Count: Total number of nodes in the cluster
  • Status Breakdown: Visual indicators for node health states
  • Quick Stats: Ready vs. total node count display
  • Health Overview: Overall node fleet health indication

Running Pods Card​

  • Pod Count: Total number of running pods across all nodes
  • Status Distribution: Pod state breakdown (Running, Pending, Failed)
  • Health Indicators: Overall pod health across the cluster
  • Capacity Information: Pod allocation efficiency

Resource Utilization Display​

CPU Utilization​

  • Overall Percentage: Cluster-wide CPU usage as a percentage
  • Visual Progress Bar: Color-coded utilization indicator
    • Green (0-70%): Healthy utilization range
    • Orange (70-85%): Warning range, monitor closely
    • Red (85%+): Critical range, action required
  • Usage Details: Current usage vs. total allocatable capacity
  • Trend Indication: Visual indicators for usage direction

![Figure needed]

Screenshot of CPU utilization display with color-coded progress bar

Memory Utilization​

  • Overall Percentage: Cluster-wide memory usage as a percentage
  • Visual Progress Bar: Same color coding scheme as CPU
  • Memory Details: Used vs. total available memory
  • Unit Conversion: Automatic unit display (GB, MB, etc.)
  • Capacity Tracking: Available vs. allocated memory breakdown

![Figure needed]

Screenshot of memory utilization display with detailed breakdown

Data Accuracy and Timing​

Real-time Data Collection​

  • Source: Data collected from Kubernetes metrics server
  • Processing: Real-time calculation of cluster-wide statistics
  • Aggregation: Intelligent aggregation across all cluster nodes
  • Validation: Data validation and error checking

Update Frequency​

  • Metrics Collection: Kubernetes metrics updated every 15 seconds
  • Dashboard Refresh: Interface updates every 30 seconds
  • Data Latency: Typical 15-45 second delay from actual usage
  • Accuracy: High accuracy for capacity planning and monitoring

Dashboard Interface Elements​

Layout Structure​

  • Header Section: Summary cards and quick statistics
  • Main Content: Detailed metrics and analytics
  • Navigation: Integrated with cluster navigation structure
  • Footer: Status information and last update timestamp

Interactive Elements​

  • Clickable Cards: Summary cards may link to detailed views
  • Hover Information: Additional details on mouse hover
  • Status Tooltips: Explanatory tooltips for status indicators
  • Refresh Controls: Manual refresh and status controls

Visual Design​

  • Color Consistency: Consistent color scheme throughout
  • Progress Indicators: Visual representation of utilization
  • Status Icons: Intuitive iconography for different states
  • Typography: Clear, readable font sizes and hierarchy

Data Interpretation​

Understanding Utilization Percentages​

CPU Utilization Ranges:
0-30%: Low utilization, potential for consolidation
30-70%: Optimal utilization range
70-85%: High utilization, monitor for scaling needs
85%+: Critical utilization, immediate action required

Memory Utilization Ranges:
0-40%: Low utilization, potential for consolidation
40-75%: Good utilization range
75-90%: High utilization, plan for scaling
90%+: Critical utilization, urgent scaling needed

Resource Planning Guidelines​

  • Scale Up Triggers: CPU over 80% or Memory over 85% sustained
  • Scale Down Opportunities: CPU under 30% and Memory under 40% sustained
  • Optimization Targets: 50-70% CPU, 60-80% Memory utilization
  • Capacity Buffer: Maintain 20-30% headroom for spikes

Use Cases​

Daily Health Checks​

  1. Morning Review: Check overall cluster health
  2. Resource Status: Verify resource utilization is within normal ranges
  3. Trend Awareness: Notice any significant changes from previous day
  4. Issue Identification: Spot potential problems early

Capacity Planning​

  1. Usage Trends: Monitor resource usage patterns over time
  2. Growth Planning: Plan for additional capacity based on trends
  3. Cost Optimization: Identify opportunities for resource optimization
  4. Scaling Decisions: Make informed decisions about cluster scaling

Performance Monitoring​

  1. Bottleneck Identification: Find resource constraints
  2. Workload Analysis: Understand cluster workload characteristics
  3. Optimization Opportunities: Identify areas for improvement
  4. SLA Monitoring: Ensure resource availability meets requirements

Best Practices​

Regular Monitoring​

  1. Consistent Schedule: Check dashboard at regular intervals
  2. Trend Awareness: Look for patterns and changes over time
  3. Documentation: Record significant observations and changes
  4. Alert Setup: Configure alerts for critical thresholds

Data Utilization​

  1. Baseline Understanding: Establish normal operating ranges
  2. Seasonal Patterns: Recognize cyclical usage patterns
  3. Anomaly Detection: Quickly identify unusual behavior
  4. Correlation Analysis: Correlate metrics with application behavior

Proactive Management​

  1. Early Warning: Use metrics for early problem detection
  2. Preventive Scaling: Scale before reaching critical thresholds
  3. Resource Planning: Plan capacity changes based on data
  4. Cost Management: Optimize resource allocation for cost efficiency

Next: Learn about Node Health Monitoring for detailed node-level metrics.