Responsible for the design and implementation of a Rack Monitoring System. The system uses a Golang agent to collect server, switch, and PDU data through IPMI and SNMP, then integrates Prometheus and Grafana to provide real-time visibility into rack equipment status.

Software Engineer

Advantech | 2022/11 – Present

Rack monitoring system architecture

🚀 Core Skills & Expertise

  • ✅ Protocol Integration: In-depth research and implementation of IPMI and SNMP (MIB) protocols, customizing data collection for different brand devices.
  • ✅ Monitoring System Development: Developed high-performance Agents using Golang, integrating Prometheus time-series database.
  • ✅ Data Visualization: Designed Grafana dashboards to transform complex data center metrics into intuitive real-time charts.

💼 Key Projects & Contributions

Rack Monitoring System Development

  1. Monitoring Agent Design & Data Collection

    • #Golang, #IPMI, #SNMP, #ipmitool, #gosnmp, #pysnmp
    • Server Monitoring: Developed a Golang monitoring Agent to communicate directly with BMC via the IPMI (ipmitool) protocol, capturing server health status and sensor data in real-time.
    • Network & Power Monitoring: Conducted in-depth research on SNMP MIB (Management Information Base) files to implement data collectors for specific brands (Switch: Netgear, Cisco; PDU: Raritan). Utilized gosnmp and pysnmp packages to ensure cross-platform and cross-device compatibility.
    • Comprehensive Metrics:
      • Switch: Collected Network Traffic, Port On/Off Speed, Stacking Status, Max Speed, Health Status, Temperature, and Fan Speed.
      • PDU: Monitored environmental sensor data (Humidity, Temperature, Vibration, etc.) and Power Control (On/Off).

  2. Data Pipeline & Storage

    • #Prometheus, #TimeSeriesDB
    • Prometheus Integration: Formatted collected heterogeneous data into unified Prometheus Metrics, establishing a high-efficiency Time-series Pipeline to support high-frequency data ingestion and querying.

  3. Grafana Real-time Monitoring Dashboard

    • #Grafana, #Dashboard, #Visualization
    • Grafana Dashboard: Designed multi-dimensional dashboards to show the operational status of rack equipment in one place.
    • Anomaly Alerting: Implemented threshold-based alerting. Visual indicators and notifications help operations personnel locate anomalies faster when overheating, fan failures, or network congestion occurs.

🛠️ Tech Stack

  • Languages: Golang, Python.
  • Protocols: IPMI (ipmitool), SNMP (v2c/v3).
  • Hardware: Netgear Switch, Cisco Switch, Raritan PDU.
  • Observability: Prometheus, Grafana.