> This page is for NeMo Fabric, version dev.
> For other versions, use one of these documentation indexes:
> - Latest (v0.6.0) (default): https://docs.nvidia.com/nemo/fabric/latest/llms.txt
> - dev: https://docs.nvidia.com/nemo/fabric/dev/llms.txt
> - v0.6.0: https://docs.nvidia.com/nemo/fabric/v0.6.0/llms.txt
> - v0.5.0: https://docs.nvidia.com/nemo/fabric/v0.5.0/llms.txt
> - v0.4.0: https://docs.nvidia.com/nemo/fabric/v0.4.0/llms.txt
> - v0.3.0: https://docs.nvidia.com/nemo/fabric/v0.3.0/llms.txt
> - v0.2.0: https://docs.nvidia.com/nemo/fabric/v0.2.0/llms.txt
> - v0.1.1: https://docs.nvidia.com/nemo/fabric/v0.1.1/llms.txt
> - v0.1.0: https://docs.nvidia.com/nemo/fabric/v0.1.0/llms.txt

> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.nvidia.com/nemo/fabric/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.nvidia.com/nemo/fabric/_mcp/server.

# Struct Sample Summary

> Resource series retention, reductions, and collection quality. Resource samples remain in the referenced JSONL artifact for interval selection.

Generated from `cargo doc --no-deps -p nemo-fabric-core`.

<pre />

Resource series retention, reductions, and collection quality. Resource samples remain in the referenced JSONL artifact for interval selection.

## Fields

### `target_interval_ms: u32`

Requested cadence; actual timing is retained in sample offsets.

### `max_samples: u32`

Retention bound, including endpoints.

### `sample_count: u32`

Retained samples, including unavailable readings and endpoints.

### `dropped_samples: u64`

Samples skipped because collection fell behind or retention was full.

### `largest_gap_seconds: f64`

Maximum observed interval between retained readings.

### `collection_seconds: f64`

Total time spent collecting retained observations, including endpoints.

### `truncated: bool`

Whether collection or retention prevented a complete series.

### `diagnostics: Vec<String>`

Safe collection/persistence diagnostics. They never change agent status.

### `samples: Option<ArtifactRef>`

Series artifact, absent when its file could not be created or fully written.

### `cgroup_target_continuous: bool`

Whether retained readings kept one cgroup identity with nondecreasing counters and no gaps over twice cadence. This does not prove exclusivity.

### `resident_memory_bytes: Option<GaugeSummary>`

Adapter-host tree resident-memory reductions.

### `process_count: Option<GaugeSummary>`

Adapter-host tree live-process-count reductions.

### `cpu_time_seconds_estimate: Measurement<f64>`

Estimated CPU counter delta, invalidated by observed resets, missing endpoints, target changes, or collection gaps larger than twice cadence.

### `cpu_percent_one_core: Measurement<f64>`

100% means one fully occupied CPU core; this is not normalized by quota.

## Trait Implementations

### `impl Clone for SampleSummary`

<pre />

#### `clone`

<pre />

#### `clone_from`

<pre />

### `impl Debug for SampleSummary`

<pre />

#### `fmt`

<pre />

### `impl<'de> Deserialize<'de> for SampleSummary`

<pre />

#### `deserialize`

<pre />

### `impl JsonSchema for SampleSummary`

<pre />

#### `schema_name`

<pre />

#### `schema_id`

<pre />

#### `json_schema`

<pre />

#### `inline_schema`

<pre />

### `impl PartialEq for SampleSummary`

<pre />

#### `eq`

<pre />

#### `ne`

<pre />

### `impl Serialize for SampleSummary`

<pre />

#### `serialize`

<pre />

### `impl StructuralPartialEq for SampleSummary`

<pre />