Hardware
Topology-Aware Workload Scheduling with NVIDIA Topograph
**Summary:** NVIDIA Topograph discovers and normalizes cluster topology from cloud APIs or on‑prem fabric systems into a common model that is published as Kubernetes node labels, Slurm configuration,
Summary: NVIDIA Topograph discovers and normalizes cluster topology from cloud APIs or on‑prem fabric systems into a common model that is published as Kubernetes node labels, Slurm configuration, or Slinky ConfigMaps, enabling work‑load managers to place AI workloads in the most efficient locality domains. It continuously updates its view of topology changes so schedulers remain current without manual intervention and supports major cloud providers (Google Cloud, Lambda, Nebius, Nscale, Oracle) as well as on‑prem InfiniBand, Spectrum-X, and Multi‑Node NVLink networks. Operators can deploy Topograph via Helm on Kubernetes or native packages on Slurm clusters, with simulation utilities available for testing; it integrates with KAI Scheduler for topology‑aware gang scheduling across AI factory infrastructure.
Related
- Achieving Peak System and Workload Efficiency on NVIDIA GB200 NVL72 with Slurm Block Scheduling
- Scaling Federated Learning Across Docker, Kubernetes, and Slurm with NVIDIA FLARE
- Running Large-Scale GPU Workloads on Kubernetes with Slurm
Source: NVIDIA Developer | 2026-09-22