Overview
System Health
Compute, inference and solver health across all production regions.
p95 inference latency
42.8ms−3.1%
GPU utilization
78.4%+4.2%
Requests / sec
12.4k+11%
Model Governance Audits
100%0.0%
Inference latency
last 24h · ms
Services
30d uptime
- Compliance auditor100.00%
- Inference gateway99.97%
- Model telemetry stream99.94%
- RAG vector database99.98%
- Data ingestion (Kafka)maint.
Accelerator fleet
5 nodes
| Node | Status | Util | Memory | Temp |
|---|---|---|---|---|
| aws-ec2-p4de-us-east-1a | hot | 92% | 71% | 74°C |
| aws-ec2-p4de-us-east-1b | healthy | 81% | 64% | 66°C |
| aws-ec2-p5-eu-west-1a | healthy | 74% | 58% | 61°C |
| aws-ec2-p5-eu-west-1b | offline | 0% | 0% | 34°C |
| aws-trainium-trn1-us-west-2 | healthy | 66% | 49% | 55°C |