Rafay Systems and Aviz Networks Partner to Optimize GPU Cloud Infrastructure

Partnership Enables Enterprises and Graphics Processing Unit (GPU) Cloud Providers to Rapidly Operationalize Multi-Tenant Artificial Intelligence (AI) Fabrics with Self-Service and Network Visibility

SAN JOSE, Calif. & SUNNYVALE, Calif. - 3 Nov, 2025 - Rafay Systems and Aviz Networks announced a strategic partnership focused on enhancing GPU cloud infrastructure consumption. This collaboration aims to streamline the adoption and monetization of GPU resources.

The partnership provides a full-stack orchestration solution that integrates Rafay's Kubernetes and GPU lifecycle management with Aviz's AI-optimized fabric orchestration and tenant-aware network automation. This combined approach allows for secure and efficient consumption of GPU resources through self-service workflows that encompass both computing and networking.

Key features of the solution include:

• End-to-End Self-Service: This feature enables developers and data scientists to access GPU/CPU resources through secure, tenant-aware network automation.
• AI-Ready Fabric Orchestration: Aviz ONES automatically configures Spectrum-X switches, GPU NICs, and server networking, enhancing east-west performance and workload isolation.
• Multi-Tenancy at Scale: Rafay handles tenant and GPU binding for Kubernetes clusters, while Aviz ensures the mapping of GPUs and NICs to logical network segments for improved isolation and visibility.
• Full Stack Observability (FSO): This provides real-time insights into both compute and network layers, facilitating quicker troubleshooting and optimizing GPU investment returns.
• Rapid Time-to-Market: Deployment timelines for GPU clouds can be reduced to weeks, as integrated APIs replace traditional manual provisioning workflows.

Haseeb Budhani, CEO and Co-Founder of Rafay Systems, stated, “Cloud providers and enterprises need a simple way to consume GPU infrastructure without reinventing orchestration stacks. Our partnership with Aviz gives customers not just self-service compute, but the tools and visibility they need to run AI workloads at scale.”

Vishal Shukla, CEO and Co-Founder of Aviz Networks, added, “Aviz was founded to make AI networking simple, open, and multi-vendor. Together with Rafay, we deliver a powerful combination that enables GPU cloud providers to achieve AWS-like efficiency with a simple, intuitive stack.”

Investment in GPU infrastructure has surged as enterprises enhance their AI capabilities. However, many existing GPU clouds face challenges related to multi-tenancy, orchestration, and visibility, leading to increased costs and underutilized resources. The partnership between Rafay and Aviz addresses these issues by integrating compute orchestration with AI fabric control, enabling more fluid consumption models in multi-vendor environments.