Nodes online
Total GPUs
Active alerts
Critical
GPUs available
Integrations
AI insights
Most recent findings across the fleet
Cluster utilization
GPU compute utilization over last 60 minutes, per cluster
Fleet health
Distribution by node status
Fleet availability
—%online right now
0 nodes not reporting
Interconnect throughput
NVLink + fabric, aggregate GB/s
GPU memory pressure
HBM used across fleet
Recent incidents
Latest 5 active alerts
Clusters
Regional distribution and headline metrics
Hottest nodes
Average GPU temperature, top 6