nvidia-resiliency-ext

Documentation contents:

  • Release Notes
  • InJob restart
    • InJob restart usage guide
    • Integration guides
    • Examples
    • Configuration and timeouts
    • Multinode deployment
    • Operations and troubleshooting
    • Optional health checks and attribution
    • Timeout dumps
    • API documentation
      • Config
      • Client
      • Server
  • Async Checkpointing
  • Local Checkpointing
  • Straggler Detection
  • Failure Attribution
  • Shared Utilities
nvidia-resiliency-ext
  • InJob restart
  • API documentation
  • View page source

API documentation

API documentation

  • Config
    • FaultToleranceConfig
      • FaultToleranceConfig.from_args()
      • FaultToleranceConfig.from_kwargs()
      • FaultToleranceConfig.is_progress_tracking_enabled
  • Client
    • RankMonitorClient
      • RankMonitorClient.calculate_and_set_hb_timeouts()
      • RankMonitorClient.calculate_and_set_section_timeouts()
      • RankMonitorClient.end_all_sections()
      • RankMonitorClient.end_section()
      • RankMonitorClient.init_workload_monitoring()
      • RankMonitorClient.load_state_dict()
      • RankMonitorClient.send_heartbeat()
      • RankMonitorClient.send_workload_control_request()
      • RankMonitorClient.shutdown_workload_monitoring()
      • RankMonitorClient.start_section()
      • RankMonitorClient.state_dict()
    • RankMonitorClientError
  • Server
    • RankMonitorLogger
      • RankMonitorLogger.log_restarter_event()
    • RankMonitorServer
Previous Next

© Copyright 2024, NVIDIA Corporation.

Built with Sphinx using a theme provided by Read the Docs.