Monitoring

Overview#

Edge Network provides built-in monitoring for all VMs at no additional cost. View real-time and historical metrics for CPU, memory, disk and network to understand your VM’s performance.

Accessing metrics#

View your VM’s metrics from the Metrics tab in the VM dashboard.

1. Navigate to your VM in the console.
2. Click the Metrics tab.
3. Select a time range (1 hour, 6 hours, 24 hours, 7 days or 30 days).

CPU metrics#

Monitor CPU utilisation to understand your VM’s processing load.

CPU usage (%)#

The percentage of CPU capacity being used, averaged across all vCPUs.

• 0–60%: normal operation
• 60–85%: increased load, monitor closely
• 85–100%: high load, consider upgrading

Memory metrics#

Track RAM usage to ensure your applications have enough memory.

Memory usage#

• Used: RAM actively in use by processes
• Cached: RAM used for disk caching (can be freed)
• Available: RAM available for new processes
• Total: total RAM allocated to your VM

Disk metrics#

Monitor disk usage and I/O performance.

• Disk usage: percentage of disk space used. Keep it below 90% to avoid performance issues.
• Disk I/O (read/write): data transfer rate to and from disk, measured in MB/s or IOPS.
• Disk latency: time for disk operations to complete. Lower is better (<10ms is excellent).

Network metrics#

Track network traffic to and from your VM.

• Network in: data received by your VM (inbound traffic), measured in Mbps.
• Network out: data sent from your VM (outbound traffic), measured in Mbps.
• Packets: number of network packets transmitted and received.

Time ranges#

Select different time ranges to view historical data:

Quick stats#

The VM Overview page shows current resource usage at a glance.

These metrics update automatically and show your VM’s current state.

Monitoring best practices#

• Check metrics regularly: review your metrics at least weekly to catch trends before they become problems.
• Understand your baseline: know what “normal” looks like for your application so you can spot anomalies.
• Scale before you need to: if resources are consistently at 70–80% utilisation, consider upgrading before issues occur.
• Correlate metrics: high CPU with normal memory often means a CPU-bound workload. High memory with CPU spikes may indicate memory pressure.

Coming soon#

• Alerts: get notified when metrics exceed thresholds
• Export: download metrics data as CSV
• Integrations: send metrics to external monitoring tools