perfsonar-mcp

Instructions

Implementation Reference

• src/perfsonar_mcp/fastmcp_server.py:156-174 (handler)

  The MCP tool handler for 'get_throughput' which calls the perfsonar_client and formats the result as a JSON string.

  async def get_throughput(
      source: str,
      destination: str,
      timeRange: int = 86400,
      summaryWindow: Optional[int] = None,
  ) -> str:
      """Get throughput measurements between source and destination.
  
      Args:
          source: Source host/IP address
          destination: Destination host/IP address
          timeRange: Time range in seconds (default: 86400 = 24 hours)
          summaryWindow: Summary window in seconds for aggregation
  
      Returns:
          JSON string with throughput measurement data
      """
      results = await perfsonar_client.get_throughput(source, destination, timeRange, summaryWindow)
      return json.dumps([r.model_dump(by_alias=True) for r in results], indent=2)

• src/perfsonar_mcp/client.py:123-150 (handler)

  The underlying client implementation that fetches the actual throughput measurement data.

  async def get_throughput(
      self,
      source: str,
      destination: str,
      time_range: Optional[int] = None,
      summary_window: Optional[int] = None,
  ) -> List[MeasurementResult]:
      """
      Get throughput measurements between source and destination
  
      Args:
          source: Source host/IP address
          destination: Destination host/IP address
          time_range: Time range in seconds from now
          summary_window: Summary window in seconds
  
      Returns:
          List of measurement results
      """
      logger.info(f"Getting throughput: {source} -> {destination}")
      metadata = await self.query_measurements(
          MeasurementQueryParams(source=source, destination=destination, event_type="throughput")
      )
  
      results = []
      for meta in metadata:
          event_type = next((e for e in meta.event_types if e.event_type == "throughput"), None)
          if not event_type: