> ## Documentation Index
> Fetch the complete documentation index at: https://docs.usechamber.io/llms.txt
> Use this file to discover all available pages before exploring further.

# Standalone Agent

> Monitor GPU utilization and workloads on bare-metal or VM-based GPU hosts without Kubernetes

<Warning>
  The Standalone Agent is currently in **beta** and supports **metric collection only**. Features and configuration may change.
</Warning>

The Chamber Standalone Agent runs on individual GPU hosts (bare metal, cloud VMs, or on-prem servers) and reports GPU, host, and application-level metrics to the Chamber dashboard. It is designed for environments where GPUs are not managed by Kubernetes.

## When to Use the Standalone Agent

| Environment                       | Agent                               |
| --------------------------------- | ----------------------------------- |
| Kubernetes cluster with GPU nodes | [Kubernetes Agent](/agent/overview) |
| Bare-metal GPU servers            | **Standalone Agent**                |
| Cloud GPU VMs (GCP, AWS, Azure)   | **Standalone Agent**                |
| On-prem GPU workstations          | **Standalone Agent**                |

## What It Does

<CardGroup cols={2}>
  <Card title="GPU Metrics" icon="microchip">
    Collects utilization, memory, temperature, power draw, and profiling-level metrics from every GPU on the host via DCGM Exporter
  </Card>

  <Card title="Host Metrics" icon="server">
    Reports CPU usage, memory usage, and capacity so you can see the full picture alongside GPU data
  </Card>

  <Card title="Workload Discovery" icon="magnifying-glass">
    Automatically detects GPU processes (PyTorch, DeepSpeed, vLLM, etc.), groups distributed training workers, and tracks their lifecycle
  </Card>

  <Card title="OTLP Receiver" icon="plug">
    Accepts OpenTelemetry metrics from your applications over gRPC, forwarding them to the Chamber dashboard alongside infrastructure metrics
  </Card>
</CardGroup>

## Architecture

```mermaid theme={null}
flowchart LR
    subgraph host["Your GPU Host"]
        Agent["Chamber<br/>Standalone Agent"]
        DCGM["DCGM Exporter"]
        GPU["NVIDIA GPUs"]
        App["Your App<br/>(OTLP)"]
    end

    DCGM -->|scrape metrics| Agent
    GPU --> DCGM
    App -->|"gRPC :4317"| Agent
    Agent -->|"WSS :443<br/>outbound only"| Chamber["Chamber<br/>Control Plane"]
```

The agent connects outbound to Chamber over a secure WebSocket. No inbound ports or firewall rules are required for Chamber communication.

## What You'll See in the Dashboard

Once the agent is running, navigate to **Dashboard > Services** ([app.usechamber.io/dashboard?tab=services](https://app.usechamber.io/dashboard?tab=services)) to see:

* **GPU utilization and GPU Usage** per host and per GPU — identify underutilized or overloaded GPUs
* **GPU memory consumption** — track memory pressure across your fleet
* **Temperature and power draw** — spot thermal throttling or power-capped GPUs before they impact training
* **Host CPU and memory** — correlate host-level bottlenecks with GPU performance
* **Active workloads** — see which processes are running on each GPU, how long they've been active, and how much GPU memory they're consuming
* **Application metrics** — if you instrument your app with OpenTelemetry, custom metrics appear alongside infrastructure data

<Info>
  Metrics are collected every 30 seconds and uploaded in batches every 60 seconds. Your host should appear in the dashboard within 60 seconds of the agent starting.
</Info>

## Beta Limitations

* **Metric collection only** — the standalone agent does not support scheduling, capacity pool management, or workload orchestration
* **No auto-remediation** — unlike the Kubernetes agent, the standalone agent does not restart or reschedule failed workloads
* **Features may change** — configuration, metric names, and dashboard views are subject to change during the beta period

## Next Steps

<CardGroup cols={2}>
  <Card title="Installation" icon="download" href="/agent/standalone/installation">
    Install the agent on a GPU host
  </Card>

  <Card title="Metrics Reference" icon="chart-line" href="/agent/standalone/metrics">
    Understand the metrics collected and what they mean
  </Card>

  <Card title="OTLP Integration" icon="plug" href="/agent/standalone/otel-integration">
    Send application-level OpenTelemetry metrics to Chamber
  </Card>

  <Card title="Configuration" icon="gear" href="/agent/standalone/configuration">
    Environment variables, workload labels, upgrading, and uninstalling
  </Card>
</CardGroup>