Workload API¶
Workload API lets users run containerized Workloads on DataRobot infrastructure. Users supply a pre-built container image or the source code to build one from; the platform handles scheduling, lifecycle management, inference routing, and observability.
The table below lists the topics included in the Workload API section.
| トピック | 説明 |
|---|---|
| Enable the Workload API | Prerequisites, deployment, feature flags and licensing, and verification. |
| Build Workloads from source | Enable code-to-workload image builds—registry and storage, network requirements, and verification. |
| Capacity & resources | Control-plane sizing, workload quotas, and CPU/GPU bundles. |
| Networking & security | Image URI validation, network policies, and workload service accounts. |
| Autoscaling & operations | KEDA autoscaling, the OpenTelemetry endpoint, and graceful shutdown. |
| Upgrade the Workload API | Data migration and CRD-version handling. |
| トラブルシューティング | Diagnose control-plane services and individual workloads. |