Netris Announces Integration and Support for NVIDIA DSX Air

Audio By Carbonatix

SANTA CLARA, Calif.--(BUSINESS WIRE)--Mar 16, 2026--

GTC 2026 —Netris, the leading provider of network automation and multi-tenancy for AI infrastructure, today announced integration and support for NVIDIA DSX Air, the infrastructure simulation platform announced by NVIDIA at GTC 2026.

NVIDIA DSX Air enables teams to simulate large-scale AI infrastructure at full fidelity, including network orchestration, automation, and multi-tenancy; Kubernetes orchestration; storage, security, observability, and other ecosystem partner solutions. The platform runs at scale, allowing teams to model complex, multi-tenant AI environments that mirror real-world deployments.

Netris deploys on NVIDIA DSX Air as the network orchestration, automation, and multi-tenancy layer across the entire AI networking stack: Ethernet (including NVIDIA Spectrum-X Ethernet ), NVL72, NVIDIA BlueField DPUs, and Virtual and Edge Networking. Netris enables operators to enforce hard multi-tenancy with hardware-level isolation, provision tenants with instant network isolation, and dynamically reallocate GPU capacity across tenants. All of this can now be developed, tested, and validated before hardware is delivered.

The integration prepares teams across the full deployment lifecycle from ISV integration and customer evaluation, through pre-deployment validation (day 0), to accelerated go-live (day 1) and ongoing change management (day 2), accelerating time to monetization and reducing deployment risk.

Why Simulation Matters

AI is driving the largest infrastructure deployment the world has ever seen. Thousands of organizations are designing, ordering, and commissioning GPU clusters simultaneously. These deployments require multiple technology components, such as network orchestration, automation, multi-tenancy, Kubernetes orchestration, storage, security, observability, and other ecosystem partner solutions that all need to be integrated with NVIDIA hardware and software and with each other.

The traditional technology rollout lifecycle of demo, trial, validation, and pre-deployment can no longer depend on physical labs. Labs are too limited in scale and typically booked, with long queues of AI operators waiting. At the same time, operators who have already ordered hardware and are waiting for delivery cannot spend precious time learning on their own equipment. They need to evaluate and validate the technology before the hardware arrives, so once the expensive hardware is there, it goes live immediately.

Netris in the AI Infrastructure Stack

Modern AI infrastructure is not a single network. It consists of multiple independent networking fabrics operating simultaneously: North-South Ethernet fabrics with and without DPUs, East-West back-end fabrics such as NVIDIA Spectrum-X Ethernet for GPU-to-GPU communication, and rack-scale NVLink interconnect domains. Each uses distinct technologies and control mechanisms, yet they must be orchestrated, validated, and operated as a cohesive system. Netris provides this orchestration.

Netris is a critical component of this stack, providing network automation, abstraction, and multi-tenancy across all of these independent traffic fabrics. Netris deploys as a complete system inside DSX Air. The same Netris controller that runs inside DSX Air moves directly into the live environment at deployment. If discrepancies exist between the digital twin and physical hardware — wrong cabling, topology mismatches — Netris immediately detects them and guides engineers on the ground to fix the issues.

Netris is the leading provider of network automation and multi-tenancy for AI infrastructure. The Netris NAAM (Network Automation, Abstraction, and Multi-Tenancy) platform is purpose-built for GPU clouds and AI Factories, trusted by high-growth neoclouds, sovereign AI cloud providers, and leading system integrators worldwide. Netris captured 12 percent of neoclouds worldwide in the last 10 months, with every deployment live across 20+ data centers.

How Customers and Partners Use DSX Air Across the Deployment Lifecycle

ISV Integration. AI deployments require multiple technology components that need integrations with NVIDIA and with each other. NVIDIA and Netris use DSX Air with ecosystem partners to develop and validate these integrations ahead of customer deployments.

Evaluation and Proof of Concept. AI operators evaluating solutions want to see demos, deep-dive into configurations, and kick the tires. Assembling physical hardware and tens of thousands of cables for an evaluation is not practical. DSX Air simulates realistic environments in minutes. ISVs and evaluating customer engineers collaborate inside DSX Sim instead of waiting months for physical lab availability.

Training Before Hardware Delivery. Once an operator signs a deal for hardware and software, they use DSX Air to stand up a simulation of their upcoming AI infrastructure running the exact anticipated software stack. Teams learn the system, ask questions to vendors ahead of time, and discover issues before hardware arrives, so once the mission-critical hardware is delivered, they go live and start making money sooner.

Pre-Deployment (Day 0). Complex AI infrastructure is deployed by skilled implementation teams that work together. These teams collaborate and validate anticipated configurations before hardware delivery. An exact, detailed digital twin is deployed in DSX Air, and implementation teams collaborate on a realistic simulation to identify issues in advance. Once hardware is delivered and powered on, the Netris controller moves from DSX Air into the live environment.

Go-Live (Day 1). Equipment powers on and teams execute with pre-validated configurations. Bring-up cycles compress because the full ecosystem has already been exercised together in simulation.

Ongoing Operations (Day 2). AI operators are live and making money. However, there are always day-2 changes — software upgrades, topology extensions, tenant policy changes, new ideas to experiment with. Instead of testing on live hardware, which should be running paid customer workloads, operators use a digital twin of their infrastructure in DSX Air to rehearse changes before applying them to the live environment.

"When mission-critical GPU hardware arrives, it should go live ASAP and not sit idle while engineers figure things out," said Alex Saroyan, CEO of Netris. "DSX Air allows our customers and partners to develop, integrate, and validate the complete ecosystem before systems are on site. With Netris, operators can instantly provision tenants, reallocate GPU capacity, and scale their business from day one, with confidence that the entire stack works in concert."

“The race to build AI factories requires a shift from manual configuration to a fully simulated, automated lifecycle that ensures the network is ready the moment GPUs are powered on,” said Amit Katz, VP of Networking at NVIDIA. “Netris integration with NVIDIA DSX Air provides a powerful environment to architect and validate the entire networking stack in parallel with hardware delivery, drastically accelerating time to value.”

Availability

Netris integration with NVIDIA DSX Air is available now. Netris will demonstrate the integration at NVIDIA GTC 2026, March 16–19 in San Jose, California. For more information, visit netris.io or contact [email protected].

About Netris

Netris is the leading provider of network automation and multi-tenancy for AI infrastructure. The Netris NAAM (Network Automation, Abstraction, and Multi-Tenancy) is the most widely deployed platform — trusted by high-growth neoclouds, NVIDIA Cloud Partners, AI factories, and leading AI platform providers. Netris provides a unified control plane and native integrations across the complete AI infrastructure networking stack — Ethernet (including NVIDIA Spectrum-X), NVIDIA Quantum InfiniBand, NVL72, NVIDIA BlueField DPUs, and virtual and edge networking. Netris enables operators to get GPU cloud business operational in weeks instead of years, provision tenants immediately with hard network isolation configured automatically, maximize GPU utilization by dynamically reallocating capacity across tenants, ensure network stability, and future-proof AI infrastructure. Learn more at netris.io.

View source version on businesswire.com:https://www.businesswire.com/news/home/20260316291228/en/

CONTACT: Media Contact:

Contact: Jason Klein

Email:[email protected]

Phone: +1 650-457-0097

KEYWORD: UNITED STATES NORTH AMERICA CALIFORNIA

INDUSTRY KEYWORD: DATA MANAGEMENT TECHNOLOGY SOFTWARE NETWORKS ARTIFICIAL INTELLIGENCE INTERNET HARDWARE

SOURCE: Netris

Copyright Business Wire 2026.

PUB: 03/16/2026 04:35 PM/DISC: 03/16/2026 04:36 PM

http://www.businesswire.com/news/home/20260316291228/en