Nvidia SMI and OSI PI Integration

Input and output integration overview

The Nvidia SMI Plugin enables the retrieval of detailed statistics about NVIDIA GPUs attached to the host system, providing essential insights for performance monitoring.

This setup converts Telegraf into a lightweight PI Web API publisher, letting you push any Telegraf metric into the OSI PI System with a simple HTTP POST.

Integration details

Nvidia SMI

The Nvidia SMI Plugin is designed to gather metrics regarding the performance and status of NVIDIA GPUs on the host machine. By leveraging the capabilities of the nvidia-smi command-line tool, this plugin pulls crucial information such as GPU memory utilization, temperature, fan speed, and various performance metrics. This data is essential for monitoring GPU health and performance in real-time, particularly in environments where GPU performance directly impacts computing tasks, such as machine learning, 3D rendering, and high-performance computing. The plugin provides flexibility by allowing users to specify the path to the nvidia-smi binary and configure polling timeouts, accommodating both Linux and Windows systems where the nvidia-smi tool is commonly located. With its ability to collect detailed statistics on each GPU, this plugin becomes a vital resource for any infrastructure relying on NVIDIA hardware, facilitating proactive management and performance tuning.

OSI PI

OSI PI is an data management and analytics platform used in energy, manufacturing, and critical infrastructure. The PI Web API is its REST interface, exposing endpoints such as /piwebapi/streams/{WebId}/value that accept JSON payloads containing a Timestamp and Value. By pairing Telegraf’s flexible HTTP output with this endpoint, any metric Telegraf collects—SNMP counters, Modbus readings, Kubernetes stats—can be written directly into PI without installing proprietary interfaces. The configuration above authenticates with Basic or Kerberos, serializes each batch to JSON, and renders a minimal body template that aligns with PI Web API’s single-value write contract. Because Telegraf already supports batching, TLS, proxies, and custom headers, this approach scales from edge gateways to cloud VMs, allowing organizations to back-fill historical data, stream live telemetry, or mirror non-PI sources (e.g., Prometheus) into the PI data archive. It also sidesteps older SDK dependencies and enables hybrid architectures where PI remains on-prem while Telegraf agents run in containers or IIoT devices.

Configuration

Nvidia SMI

[[inputs.nvidia_smi]]
  ## Optional: path to nvidia-smi binary, defaults "/usr/bin/nvidia-smi"
  ## We will first try to locate the nvidia-smi binary with the explicitly specified value (or default value),
  ## if it is not found, we will try to locate it on PATH(exec.LookPath), if it is still not found, an error will be returned
  # bin_path = "/usr/bin/nvidia-smi"

  ## Optional: timeout for GPU polling
  # timeout = "5s"

OSI PI

[[outputs.http]]
  ## PI Web API endpoint for writing a single value to a PI Point by Web ID
  url = "https://${PI_HOST}/piwebapi/streams/${WEB_ID}/value"

  ## Use POST for each batch
  method       = "POST"
  content_type = "application/json"

  ## Basic-auth header (base64-encoded "DOMAIN\\user:password")
  headers = { Authorization = "Basic ${BASIC_AUTH}" }

  ## Serialize Telegraf metrics as JSON
  data_format           = "json"
  json_timestamp_units  = "1ms"

  ## Render the JSON body that PI Web API expects
  body_template = """
  {{ range .Metrics -}}
  { "Timestamp": "{{ .timestamp | formatDate \"2006-01-02T15:04:05Z07:00\" }}", "Value": {{ index .fields 0 }} }
  {{ end -}}
  """

  ## Tune networking / batching if needed
  # timeout     = "10s"
  # batch_size  = 1

Input and output integration examples

Nvidia SMI

1. Real-Time GPU Monitoring for ML Training: Continuously monitor the GPU utilization and memory usage during machine learning model training. This enables data scientists to ensure that their GPUs are not being overutilized or underutilized, optimizing resource allocation and reviewing performance bottlenecks in real-time.

2. Automated Alerts for Overheating GPUs: Implement a system using the Nvidia SMI plugin to track GPU temperatures and set alerts for instances where temperatures exceed safe thresholds. This proactive monitoring can prevent hardware damage and improve system reliability by alerting administrators to potential cooling issues before they result in failure.

3. Performance Baselines for GPU Resources: Establish baseline performance metrics for your GPU resources. By regularly collecting data and analyzing trends in GPU usage, organizations can identify anomalies and optimize their workloads accordingly, leading to enhanced operational efficiency.

4. Dockerized GPU Usage Insights: In a containerized environment, use the plugin to monitor GPU performance from within a Docker container. This allows developers to track GPU performance of their applications in production, facilitating troubleshooting and performance optimization within isolated environments.

OSI PI

1. Remote Pump Stations Telemetry Bridge: Install Telegraf on edge gateways at oil-field pump stations, gather flow-meter and vibration readings over Modbus, and POST them to the PI Web API. Operations teams view real-time data in PI Vision without deploying heavyweight PI interfaces, while bandwidth-friendly batching keeps satellite links economical.

2. Green-Energy Micro-Grid Dashboard: Export inverter, battery, and weather metrics from MQTT into Telegraf, which relays them to PI. PI AF analytics can calculate real-time power balance and feed a campus dashboard; historical deltas inform sustainability reports.

3. Brownfield SCADA Modernization: Legacy PLCs logged to CSV are ingested by Telegraf’s tail input; each row is parsed and immediately sent to PI via HTTP, creating a live data stream that co-exists with archival files while the SCADA upgrade proceeds incrementally.

4. Synthetic Data Generator for Training: Telegraf’s exec input can run a script that emits simulated sensor patterns. Posting those metrics to a non-production PI server through the Web API supplies realistic datasets for PI Vision training sessions without risking production tags.

Feedback

Thank you for being part of our community! If you have any general feedback or found any bugs on these pages, we welcome and encourage your input. Please submit your feedback in the InfluxDB community Slack.