Foundation component catalog
Implementation detail for shared BullSequana AI platform components.
Use component pages for deployment, operations, and troubleshooting. Start with the Foundation capability overview for product-oriented guidance.
Delivery and policy
External delivery integrations
Identity and secrets
Networking
Inference
Observability
Storage, databases, and workflows
Platform internals
Components described on other pages
These Foundation components have no page of their own. Their behavior is described where it matters:
| Component | Role | Described in |
|---|---|---|
| KServe | Model-serving control plane | KServe and Inference |
| Internal model gateway | Private gateway between the BSQAI API and model workloads | Inference and Networking |
| Envoy AI Gateway and Gateway API Inference Extension | Private inference data path and model-aware routing | Inference |
| NVIDIA GPU Operator | GPU discovery, device plugin, and runtime class on GPU nodes | Enable GPU for AI components |
| Alloy OTLP gateway and telemetry query gateway | Tenant routing of application telemetry and authorization of Grafana queries | Observability and audit |