NVIDIA Run:ai GPU partitioning validated on Amazon EKS
The story so far
NVIDIA has released the GPU Fractioning Operator, which integrates scheduling and memory contracts per Pod for GPU memory sharing on Amazon EKS using Kubernetes, supporting GPUs without MIG. In the validation example, a single A10G GPU can allocate 6 GiB of memory to two Pods and supports time-slicing or sm-sharing modes.
AI-written from the reports · Updated 7 hours ago
Coverage timeline
Follow the coverage to see every side of the story.
- AWS Machine Learning BlogUsing NVIDIA Run:ai GPU Fractioning on Amazon EKS
NVIDIA Run:ai GPU partitioning on Amazon EKS achieves GPU memory sharing through Kubernetes integration scheduling and memory contracts per Pod, supporting GPUs without MIG. The validation example shows a single A10G GPU can allocate 6 GiB of memory to two Pods and supports time-slicing or sm-sharing modes.
Heat over time
- Now
- 8
- Peak
- 1020 Nov, 01:00
- 24h change
- –
Hover or tap the chart to see each hour; use the arrow keys on a keyboard.