join us at booth 1650
Rafay at Kubecon 2025
Where: Georgia World Congress Center
285 Andrew Young International Blvd. NW
Atlanta, GA 3013
Dates: November 10 - 13, 2025
Join Rafay at booth 1650 and get a demo of the Rafay Platform. Ask us about Dynamic Resource Allocation to effectively manage GPUs on Kubernetes.
- Better GPU Utilization – DRA lets NVIDIA split and share GPUs (e.g., fractional GPUs, MIG), improving efficiency and avoiding waste.
- Support for Advanced GPU Features – It standardizes allocation of GPU features like MIG profiles, SR-IOV, and accelerators across Kubernetes.
- Cloud-Native AI/ML Ecosystem – DRA enables multi-tenancy, elasticity, and fair scheduling for AI/ML workloads, making Kubernetes more GPU-friendly.
- Easier Integration with GPU Operator – NVIDIA can integrate its GPU Operator cleanly with Kubernetes scheduling instead of relying on custom hacks.
- Strategic Advantage in AI Infrastructure – By shaping DRA, NVIDIA ensures its GPUs and accelerators remain the go-to choice for cloud-native AI infrastructure.
Trusted by leading enterprises, neoclouds and service providers
Download the Reference Architecture
GPU PaaS Reference Architecture
In collaboration with NVIDIA, this reference architecture shows enterprises and cloud providers how the Rafay Platform delivers self-service consumption of AI and cloud-native infrastructure to developers and data scientists.
