AVIZ details multi-tenant networking in Spectrum-X
2 companies named across 6 categories, one of 855 articles referencing Aviz Networks. Previous coverage: Aviz details Host-Based Networking on BlueField-3 DPUs (Oct 2026).
Companies mentioned
Best suited for
- Seniority
- Architect
- Job function
- Network / Network Architect
- Persona
- Network Architect
- Buyer role
- Architect / Technical Evaluator
- Buyer journey
- Need to Buy
- Adoption curve
- Early Majority
- Technology maturity
- Operational Expansion
- Industry
- Information Technology / Networking / Enterprise & Data Center Networking / Data Center Switching & Fabrics
Our classification, not the publisher's statement. Best suited for, not only for.
AVIZ ONES adds tenant-aware networking to NVIDIA Spectrum-X by extending isolation into north-south and storage traffic. The blog says the approach is intended to help shared AI environments support multiple users and teams with consistent policy enforcement.
Research Overview
The blog describes AI infrastructure as shifting from single-tenant GPU clusters to shared platforms. In that model, GPUs, storage, and external networks serve multiple teams at the same time.
It says that as utilization rises, operators need performance, security, and simpler operations. The post positions tenant-aware networking as a response to those requirements.
Technical Breakdown
Spectrum-X is presented as an Ethernet fabric optimized for GPU east-west traffic. AVIZ ONES is described as the orchestration layer that designs, provisions, and manages the fabric as one logical network.
The blog says this extends multi-tenancy beyond GPUs into front-end and storage networks. It also states that each tenant gets end-to-end traffic isolation and consistent policy enforcement across those network paths.
Operational Impact
The post says north-south traffic connects GPU workloads to external data sources, enterprise and cloud networks, and management systems. It adds that without built-in multi-tenancy, organizations can face manual workflows, limited isolation, and scaling constraints.
By treating the fabric as shared infrastructure, the blog says organizations can support multiple teams without redesigning networks. It also says the model can raise utilization while reducing operational risk.
Leadership Perspective
The blog frames the architecture as a way to present AI infrastructure as an internal platform. It also says the fabric can integrate with cloud and PaaS ecosystems.
Its conclusion says the result is a secure, scalable foundation for multi-tenant AI data centers. This Blog Summary is a fact-based summary of the vendor blog.
Blog post, originally published at aviznetworks.com.