System Monitor

SAR · PVE01/02/03 · OLLAMA · OPENWEBUI · N8N
--:--:--
OLLAMA
PVE
N8N
WEBUI

Active Alerts

Connecting to Prometheus...

Overview

Ollama
checking...
Models in VRAM
currently loaded
Active Inference
running requests
Open WebUI
checking...
PVE Nodes Up
of 3 expected
PVE VMs Running
across cluster
Total GPU Power
watts combined
Est. Cost / Hour
@ /bin/sh.12/kWh
💲

Power Cost Estimate

P100 Draw
watts
P40 Draw
watts
Combined
watts
Per Hour
USD
Per Day
USD (24h)

GPU Telemetry — Tesla P100 & P40

Tesla P100 · 16GB HBM2
Temperature
°C / 83°C TJ
Power Draw
W / 250W TDP
Core Util
% utilized
VRAM Used
MB / 16384MB
Mem Util
% bandwidth
SM Clock
MHz
Tesla P40 · 24GB GDDR5
Temperature
°C / 83°C TJ
Power Draw
W / 250W TDP
Core Util
% utilized
VRAM Used
MB / 24576MB
Mem Util
% bandwidth
SM Clock
MHz
📈

GPU Time Series — Last 30 Min

Temperature (°C)
Power Draw (W)
Core Utilization (%)
VRAM Used (MB)
🤖

Ollama — Active Models

Model Name Status VRAM Allocated Model Size Load Events (1h) Unload Events (1h)
Loading model data...
💬

Open WebUI

Status
Active Sessions
Model Loads (1h)
Inference Running
🖥️

Proxmox VE Cluster

Loading PVE data...
📊

Proxmox — Resource Usage

CPU % — All Nodes
Memory % — All Nodes
VM Count per Node

n8n Workflow Automation — nexus.xinle.biz/n8n

n8n Workflows
Connecting...
Enter your n8n API key above and click Apply to load workflows.
🔗

Quick Access