Nvidia SMI and SQLite 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.

Telegraf’s SQL output plugin stores metrics in an SQL database by creating tables dynamically for each metric type. When configured for SQLite, it utilizes a file-based DSN and a minimal SQL schema tailored for lightweight, embedded database usage.

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.

SQLite

The SQL output plugin writes Telegraf metrics to an SQL database using a dynamic schema where each metric type corresponds to a table. For SQLite, the plugin uses the modernc.org/sqlite driver and requires a DSN in the format of a file URI (e.g., ‘file:/path/to/telegraf.db?cache=shared’). This configuration leverages standard ANSI SQL for table creation and data insertion, ensuring compatibility with SQLite’s capabilities.

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"

SQLite

[[outputs.sql]]
  ## Database driver
  ## Valid options: mssql (Microsoft SQL Server), mysql (MySQL), pgx (Postgres),
  ## sqlite (SQLite3), snowflake (snowflake.com), clickhouse (ClickHouse)
  driver = "sqlite"

  ## Data source name
  ## For SQLite, the DSN is a filename or URL with the scheme "file:".
  ## Example: "file:/path/to/telegraf.db?cache=shared"
  data_source_name = "file:/path/to/telegraf.db?cache=shared"

  ## Timestamp column name
  timestamp_column = "timestamp"

  ## Table creation template
  ## Available template variables:
  ##  {TABLE}        - table name as a quoted identifier
  ##  {TABLELITERAL} - table name as a quoted string literal
  ##  {COLUMNS}      - column definitions (list of quoted identifiers and types)
  table_template = "CREATE TABLE {TABLE} ({COLUMNS})"

  ## Table existence check template
  ## Available template variables:
  ##  {TABLE} - table name as a quoted identifier
  table_exists_template = "SELECT 1 FROM {TABLE} LIMIT 1"

  ## Initialization SQL (optional)
  init_sql = ""

  ## Maximum amount of time a connection may be idle. "0s" means connections are never closed due to idle time.
  connection_max_idle_time = "0s"

  ## Maximum amount of time a connection may be reused. "0s" means connections are never closed due to age.
  connection_max_lifetime = "0s"

  ## Maximum number of connections in the idle connection pool. 0 means unlimited.
  connection_max_idle = 2

  ## Maximum number of open connections to the database. 0 means unlimited.
  connection_max_open = 0

  ## Metric type to SQL type conversion
  ## The values on the left are the data types Telegraf has and the values on the right are the SQL types used when writing to SQLite.
  #[outputs.sql.convert]
  #  integer       = "INT"
  #  real          = "DOUBLE"
  #  text          = "TEXT"
  #  timestamp     = "TIMESTAMP"
  #  defaultvalue  = "TEXT"
  #  unsigned      = "UNSIGNED"
  #  bool          = "BOOL"

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.

SQLite

1. Local Monitoring Storage: Configure the plugin to write metrics to a local SQLite database file. This is ideal for lightweight deployments where setting up a full-scale database server is not required.
2. Embedded Applications: Use SQLite as the backend for applications embedded in edge devices, benefiting from its file-based architecture and minimal resource requirements.
3. Quick Setup for Testing: Leverage SQLite’s ease of use to quickly set up a testing environment for Telegraf metrics collection without the need for external database services.
4. Custom Schema Management: Adjust the table creation templates to predefine your schema if you require specific column types or indexes, ensuring compatibility with your application’s needs.

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.