Skip to main content

Load Balancer Resource Monitoring

The Monitoring feature in vCloud provides comprehensive charts and metrics for real-time Load Balancer performance tracking, including CPU, RAM, disk (IOPS), and network traffic. This monitoring helps assess operational status, detect early anomalies, and optimize resource usage.

Accessing Monitoring​

To access Load Balancer monitoring:

  1. Log in to the vCloud system
  2. Navigate to the Load Balancer detail page
  3. Select the Monitoring tab

Performance Monitoring Charts​

The Monitoring interface provides real-time charts for 4 key Load Balancer metrics:

Figure needed: [CPU usage chart showing percentage utilization over time with peak indicators] Figure needed: [Memory usage chart displaying RAM consumption in MB with allocation limits] Figure needed: [Network utilization chart showing inbound/outbound traffic rates in Kbps]

1. CPU Usage​

  • Chart: CPU usage as a percentage during the interval (%)
  • Unit: Percentage (%)
  • Meaning: Reflects CPU utilization against the allocated Load Balancer limit

Chart Analysis:

  • High peaks indicate periods with many concurrent connections
  • Sustained high CPU usage (>80%) may require Load Balancer resizing
  • High CPU may result from numerous SSL connections (when using HTTPS with SSL termination)

2. Memory Usage​

  • Chart: Percentage of host physical memory consumed (MB)
  • Unit: MB
  • Meaning: RAM capacity currently used by the Load Balancer

Chart Analysis:

  • RAM increases with many concurrent connections requiring maintenance
  • Load Balancer uses RAM for connection information, sessions, and caching
  • RAM approaching allocation limits may cause overload and connection rejection

3. Disk IOPS​

  • Metrics:
    • Disk Read IOPS: Number of disk reads per second
    • Disk Write IOPS: Number of disk writes per second
  • Meaning: Measures storage access traffic for the Load Balancer

Chart Analysis:

  • Load Balancers typically have lower disk read/write activity than application servers
  • High IOPS may result from logging activity, especially with detailed logging enabled
  • Sudden IOPS spikes usually relate to configuration changes or system updates

4. Network Utilization​

  • Metrics:
    • Receive-rates (download) – Kbps
    • Transmit-rates (upload) – Kbps
  • Meaning: Reflects Load Balancer inbound/outbound network traffic

Chart Analysis:

  • This is the most important metric for Load Balancers, directly reflecting traffic being processed
  • Traffic should be relatively balanced between receive and transmit for most web applications
  • Large differences between inbound and outbound traffic may indicate:
    • Data compression enabled
    • Unbalanced content distribution
    • Potential backend server issues

Monitoring Best Practices​

  • Set alerts for CPU usage above 80% sustained for 5+ minutes
  • Monitor connection patterns during peak hours to plan capacity
  • Review health check failures to identify backend server issues
  • Track SSL termination impact on CPU usage for HTTPS listeners
  • Correlate metrics between Load Balancer and backend servers for comprehensive analysis