Nvidia SMI and SigNoz 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 configuration turns any Telegraf agent into a Remote Write publisher for SigNoz, streaming rich metrics straight into the SigNoz backend with a single URL change.

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.

SigNoz

SigNoz is an open source observability platform that stores metrics, traces, and logs. When you deploy SigNoz, its signoz-otel-collector-metrics service exposes a Prometheus Remote Write receiver (default :13133/api/v1/write). By configuring Telegraf’s Prometheus plugin to point at this endpoint, you can push any Telegraf collected metrics, SNMP counters, cloud services, or business KPIs—directly into SigNoz. The plugin natively serializes metrics in the Remote Write protobuf format, supports external labels, metadata export, retries, and TLS or bearer-token auth, so it fits zero-trust and multi-tenant SigNoz clusters. Inside SigNoz, the data lands in ClickHouse tables that back Metrics Explorer, alert rules, and unified dashboards. This approach lets organizations unify Prometheus and OTLP pipelines, enables long-term retention powered by ClickHouse compression, and avoids vendor lock-in while retaining PromQL-style queries.

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"

SigNoz

[[outputs.prometheusremotewrite]]
  ## SigNoz OTEL-Collector metrics endpoint (Prometheus Remote Write receiver)
  ## Default port is 13133 when you install SigNoz with the Helm chart
  url = "http://signoz-otel-collector-metrics.monitoring.svc.cluster.local:13133/api/v1/write"

  ## Add identifying labels so you can slice & dice the data later
  external_labels = { host = "${HOSTNAME}", agent = "telegraf" }

  ## Forward host metadata for richer dashboards (SigNoz maps these to ClickHouse columns)
  send_metadata = true

  ## ----- Authentication (comment out what you don’t need) -----
  # bearer_token   = "$SIGNOZ_TOKEN"          # SaaS tenant token
  # basic_username = "signoz"                 # Basic auth (self-hosted)
  # basic_password = "secret"

  ## ----- TLS options (for SaaS or HTTPS self-hosted) -----
  # tls_ca                  = "/etc/ssl/certs/ca.crt"
  # tls_cert                = "/etc/telegraf/certs/telegraf.crt"
  # tls_key                 = "/etc/telegraf/certs/telegraf.key"
  # insecure_skip_verify    = false

  ## ----- Performance tuning -----
  max_batch_size = 10000      # samples per POST
  timeout        = "10s"
  retry_max      = 3

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.

SigNoz

1. Multi-Cluster Federated Monitoring: Drop a Telegraf DaemonSet into each Kubernetes cluster, tag metrics with cluster=<name>, and Remote Write them to a central SigNoz instance. Ops teams get a single PromQL window across prod, staging, and edge clusters without running Thanos sidecars.

2. Factory-Floor Edge Gateway: A rugged Intel NUC on the shop floor runs Telegraf to scrape Modbus PLCs and environmental sensors. It batches readings every 5 seconds and pushes them over an intermittent 4G link to SigNoz SaaS. ClickHouse compression keeps costs low while AI-based outlier detection in SigNoz flags overheating motors before failure.

3. SaaS Usage Metering: Telegraf runs alongside each micro-service, exporting per-tenant counters (api_calls, gigabytes_processed). Remote Write streams the data to SigNoz where a scheduled ClickHouse materialized view aggregates usage for monthly billing—no separate metering stack required.

4. Autoscaling Feedback Loop: Combine Telegraf’s Kubernetes input with the Remote Write output to publish granular pod CPU and queue-length metrics into SigNoz. A custom SigNoz alert fires when P95 latency breaches 200 ms and a GitOps controller reads that alert to trigger a HorizontalPodAutoscaler tweak—closing the loop between observability and automation.

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.