Network metrics are indicators that measure network performance, availability and health.

What are Network Metrics?

Network metrics are quantitative measures that provide information about network performance, availability and health.

Metric Types

Performance

  • Throughput: Transmission capacity
  • Latency: Response time
  • Jitter: Latency variation
  • Packet Loss: Packet loss

Availability

  • Uptime: Uptime
  • Downtime: Downtime
  • MTBF: Mean time between failures
  • MTTR: Mean time to repair

Utilization

  • Bandwidth: Bandwidth utilization
  • CPU: CPU utilization
  • Memory: Memory utilization
  • Storage: Storage utilization

Performance Metrics

Latency

  • RTT: Round Trip Time
  • One-way: One-way latency
  • Jitter: Latency variation
  • Baseline: Baseline latency

Throughput

  • Bits per second: Bits per second
  • Packets per second: Packets per second
  • Frames per second: Frames per second
  • Transactions per second: Transactions per second

Quality

  • Packet Loss: Packet loss
  • Error Rate: Error rate
  • Retransmissions: Retransmissions
  • Duplicates: Duplicate packets

Measurement Tools

Basic Commands

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11

# Ping for latency
ping -c 10 8.8.8.8

# Traceroute for route
traceroute 8.8.8.8

# Netstat for connections
netstat -an

# ss for connections
ss -tuln

Advanced Tools

  • iperf: Bandwidth measurement
  • nload: Traffic monitoring
  • iftop: Traffic monitoring
  • nethogs: Process monitoring

Continuous Monitoring

  • SNMP: Simple Network Management Protocol
  • NetFlow: Network flow analysis
  • sFlow: Flow sampling
  • IPFIX: Internet Protocol Flow Information Export

Application Metrics

Web

  • Response Time: Response time
  • Throughput: Processing capacity
  • Error Rate: Error rate
  • Availability: Availability

Database

  • Query Time: Query time
  • Connections: Active connections
  • Lock Time: Lock time
  • Cache Hit Rate: Cache hit rate

Email

  • Delivery Time: Delivery time
  • Bounce Rate: Bounce rate
  • Spam Rate: Spam rate
  • Queue Length: Queue length

Dashboards

Real-time Metrics

  • Grafana: Metric visualization
  • Kibana: Log analysis
  • Prometheus: Monitoring and alerts
  • Zabbix: Infrastructure monitoring

Alerts

  • Thresholds: Threshold configuration
  • Notifications: Automatic notifications
  • Escalation: Alert escalation
  • Correlation: Event correlation

Best Practices

Measurement

  • Baseline: Establish baseline
  • Regularity: Regular measurement
  • Context: Measurement context
  • Documentation: Document metrics

Analysis

  • Trends: Trend analysis
  • Correlation: Metric correlation
  • Anomalies: Anomaly detection
  • Root Cause: Root cause analysis

Optimization

  • Identification: Identify bottlenecks
  • Prioritization: Prioritize optimizations
  • Implementation: Implement improvements
  • Validation: Validate improvements
  • NPM - Network monitoring that generates metrics
  • Dashboards - Network metrics visualization
  • Logs - Logs that generate network metrics
  • Telegraf - Tool that collects network metrics
  • InfluxDB - Database that stores network metrics
  • Traffic Captures - Data that generates network metrics
  • Firewall - Device that generates network metrics
  • VPN - Connection that generates network metrics
  • VLAN - Segment that generates network metrics
  • Routers - Devices that generate network metrics
  • Switches - Devices that generate network metrics
  • CISO - Role that oversees network metrics

References