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

# Grafana Integration

> Visualize Visent GPU metrics in Grafana dashboards with pre-built templates

# Grafana Integration

Create comprehensive GPU monitoring dashboards in Grafana using Visent Telemetry metrics. Import pre-built dashboard templates, customize visualizations, and set up alerting for GPU infrastructure monitoring.

Supports both Prometheus and direct API data sources with real-time metric visualization.

## Prerequisites

Coming soon - Grafana installation and basic configuration requirements.

## Data Source Configuration

### Prometheus Data Source

```json theme={null}
{
  "name": "Visent Prometheus",
  "type": "prometheus",
  "url": "http://prometheus:9090",
  "access": "proxy",
  "isDefault": false
}
```

### Direct API Data Source

Coming soon - configuring Grafana to query Visent API directly.

## Dashboard Templates

### GPU Overview Dashboard

Coming soon - comprehensive GPU fleet overview dashboard template.

### Node-Level Dashboard

Coming soon - detailed per-node GPU monitoring dashboard.

### Workload Analytics Dashboard

Coming soon - workload-specific performance analysis dashboard.

## Custom Dashboards

### Panel Configuration

```json theme={null}
{
  "title": "GPU Utilization",
  "type": "graph",
  "targets": [
    {
      "expr": "visent_gpu_utilization",
      "legendFormat": "GPU {{gpu}} - {{node}}"
    }
  ],
  "yAxes": [
    {
      "min": 0,
      "max": 100,
      "unit": "percent"
    }
  ]
}
```

### Alert Configuration

Coming soon - setting up Grafana alerts for GPU metrics.

## Visualization Types

### Time Series Graphs

Coming soon - trending GPU performance over time.

### Heatmaps

Coming soon - GPU utilization heatmaps across clusters.

### Single Stat Panels

Coming soon - key performance indicators and current values.

### Table Panels

Coming soon - detailed metric tables and comparisons.

## Dashboard Management

### Import/Export

Coming soon - sharing and managing dashboard configurations.

### Provisioning

Coming soon - automated dashboard deployment and updates.

### Variables

Coming soon - using dashboard variables for dynamic filtering.

## Performance Optimization

Coming soon - optimizing Grafana performance for large-scale GPU monitoring.

## Troubleshooting

Coming soon - common Grafana integration issues and solutions.

## Next Steps

* [Configure Prometheus integration](prometheus) as data source
* [Set up alerts](../telemetry/alerts) for comprehensive monitoring
* [Explore API access](../api/telemetry) for custom integrations