Data Center: Resource Monitor

The Monitor page provides a real-time overview of all rendering node resources in the cluster.

![](https://docs.paraverse.cc/media/202504/49d46d33b3594f1dbddaff7da325067d3795.png)

## Features

- **GPU node summary** — Total node count, concurrent utilization, offline nodes, and alarm statuses (CPU, GPU, memory, disk).
- **Resource failure alerts** — Threshold warnings for abnormal resource consumption.
- **Node resource table** — Per-node details including node name, IP address, max concurrency, CPU usage, GPU usage, VRAM usage, memory usage, and disk usage.
- **Node management** — Batch freeze online nodes, batch delete offline nodes, and batch restart nodes (service-level or full server restart).
- **Remote control** — Connect to and control specific GPU nodes remotely.
- **1-minute usage graphs** — View fluctuation charts for CPU, GPU, VRAM, memory, disk, and network usage over the last minute.

## Query and Filter

Filter nodes by **Region ID**, **Server IP**, **Rendering Node Name**, or **Node Status**.

## Refresh Rate

Data refreshes every **5 seconds** by default. An option to refresh every **2 seconds** is available for higher-frequency monitoring.