Module resource_metrics

This page is for NeMo Relay, version dev. For other versions, use one of these documentation indexes: - Latest (v0.11.1) (default): https://docs.nvidia.com/nemo/relay/latest/llms.txt - dev: https://docs.nvidia.com/nemo/relay/dev/llms.txt - v0.11.1: https://docs.nvidia.com/nemo/relay/v0.11.1/llms.txt - v0.10.4: https://docs.nvidia.com/nemo/relay/v0.10.4/llms.txt - v0.9.4: https://docs.nvidia.com/nemo/relay/v0.9.4/llms.txt - v0.8.4: https://docs.nvidia.com/nemo/relay/v0.8.4/llms.txt - v0.7.3: https://docs.nvidia.com/nemo/relay/v0.7.3/llms.txt

Generated from cargo doc --no-deps -p nemo-relay-builtins -p nemo-relay-diagnostics -p nemo-relay-logging -p nemo-relay-plugin-host -p nemo-relay-codecs -p nemo-relay -p nemo-relay-runtime -p nemo-relay-model-pricing -p nemo-relay-resource-metrics -p nemo-relay-observability -p nemo-relay-adaptive -p nemo-relay-pii-redaction -p nemo-relay-ffi -p nemo-relay-types -p nemo-relay-plugin -p nemo-relay-worker-proto -p nemo-relay-worker.

Resource collection and supervised-process integration. Fresh and managed system resource metric observation.

Structs

• AcceleratorDeviceMetrics: Device-wide accelerator measurements.
• AcceleratorProcessMetrics: Accelerator measurements attributed to one process in the selected scope.
• CpuMetrics: CPU measurements and CPU limit events.
• DiskMetrics: Disk I/O for the measured process scope and filesystem capacity.
• FilesystemCapacityMetrics: Filesystem capacity for one configured path.
• GpuMetrics: Accelerator (GPU) device-wide and process-attributed measurements.
• MemoryMetrics: Memory measurements and memory limit events.
• NetworkInterfaceMetrics: Network traffic for one visible interface.
• NetworkMetrics: System-wide network traffic, regardless of the process measurement scope.
• NetworkTrafficMetrics: Network counters for one interface or the selected system aggregate.
• ProcessMetrics: Process counts, process limits, handles, and descriptors.
• ProcessSamplingMetadata: Coverage of process queries used for process-summed measurements.
• ResourceLimitEventCount: One typed cumulative resource-limit event counter.
• ResourceMeasurement: One available resource measurement.
• ResourceMetricsConfig: Configuration for the built-in resource_metrics plugin component.
• ResourceMetricsCpuConfig: CPU observations, limits, and pressure.
• ResourceMetricsCpuUnits: CPU measurement unit choices.
• ResourceMetricsDiskConfig: Disk I/O for the selected process scope and filesystem capacity.
• ResourceMetricsDiskUnits: Disk measurement unit choices.
• ResourceMetricsGpuConfig: Accelerator observations.
• ResourceMetricsGpuUnits: GPU measurement unit choices.
• ResourceMetricsMemoryConfig: Memory observations, limits, pressure, and OOM events.
• ResourceMetricsMemoryUnits: Memory measurement unit choices.
• ResourceMetricsNetworkConfig: System interface counters. Empty interfaces selects every visible interface.
• ResourceMetricsNetworkTrafficUnits: Unit choices for one group of network traffic.
• ResourceMetricsNetworkUnits: Unit choices for system and interface traffic.
• ResourceMetricsPollingConfig: Periodic collection controlled by the resource_metrics plugin component.
• ResourceMetricsProcessConfig: Process, thread, handle, descriptor, and process-limit observations.
• ResourceMetricsSnapshot: One point-in-time observation of the selected system and process resources.
• ResourceMetricsUnits: Output unit choices for resource metrics.

Enums

• AcceleratorVendor: Accelerator vendor associated with a metric record.
• BandwidthUnit: Units for data transferred per second.
• CapacityUnit: Units for memory and filesystem capacity.
• CountUnit: Fixed units for process, resource, and operation counts.
• CpuUnit: Units for CPU consumption and effective capacity.
• DataUnit: Units for cumulative transferred data.
• DurationUnit: Units for elapsed and cumulative durations.
• ResourceLimitEventKind: Direct operating-system resource-limit event kind.
• ResourceLimitResource: Resource whose configured boundary produced a limit event.
• ResourceMeasurementScope: System or process boundary measured by a snapshot.
• ResourceMetricValue: Exact integral values remain integers; converted fractional values are decimals.
• ResourceMetricsMeasurementScope: Measurement boundary selected for resource metric collection.
• ResourceOperatingSystem: Operating system that produced a resource metrics snapshot.
• UtilizationUnit: Units for GPU compute utilization.

Constants

• RESOURCE_METRICS_EVENT_NAME: Canonical event name for resource metric marks. Canonical event name for resource metric marks.

Traits

• ResourceUnit: A unit with a stable name in snapshots and metric marks.

Functions

• collect: Acquire a fresh snapshot without requiring managed polling.