Nscale's tests show NVIDIA DSX MaxLPS fits 37% more GPUs in the same power budget
37% more GPUs
37% more GPUs and ~50% higher inference throughput within the same power budget
European AI cloud provider Nscale tested NVIDIA DSX MaxLPS on GB300 NVL72 systems to match power allocation to workloads. The tests showed 37% more GPUs and nearly 50% higher aggregate inference throughput within the same provisioned power. Nscale plans Vera Rubin deployments across five European markets.
The tools behind this win
- NVIDIA DSX MaxLPSCoordinated power allocation with workloads to add GPUs and inference throughput
- NVIDIA GB300 NVL72Hardware platform used in the power-efficiency tests