Netris provides a network‑automation platform that abstracts AI‑centric networking fabrics—including Ethernet, InfiniBand, NVLink, and DPUs—into cloud‑grade constructs such as VPCs, elastic IPs, and load balancers. Engineers can provision VRFs, VXLANs, BGP policies and enforce hardware‑level multi‑tenancy across bare‑metal, VM, and container workloads via a web console, REST API, or Terraform provider, with built‑in validation and auto‑remediation.
Funding
$3.3M raised to dateRaised to date based on public sources. This may differ from the amount the company actually raised and is based only on what is publicly available on the internet.
Founders
Product
Problem
AI cloud operators must orchestrate networking across heterogeneous fabrics—Ethernet, InfiniBand, NVLink, and DPUs—while providing secure, hardware‑level multi‑tenancy. Existing SDN controllers are built for legacy enterprise environments and lack native support for GPU‑centric hardware, forcing operators to develop fragile in‑house automation or rely on soft isolation that does not meet compliance requirements.
Solution
Netris delivers a network‑automation platform that abstracts the full AI networking stack into cloud‑provider‑grade constructs. Engineers define desired end states (e.g., VRFs, VXLANs, BGP policies) through a web console, REST API, or Terraform provider, and the system translates them into device‑specific configurations across switches, DPUs, and fabrics in minutes. The platform enforces hard isolation at the hardware layer, guaranteeing tenant separation for bare‑metal, VM, and container workloads. Built‑in algorithms provide closed‑loop validation and auto‑remediation, reducing day‑0 deployment time and eliminating manual errors during day‑2 operations. By exposing AWS‑style services such as VPCs, elastic IPs, and load balancers, Netris enables dynamic resource sharing and rapid scaling of GPU resources without vendor lock‑in. The solution integrates with NVIDIA Spectrum switches and other supported hardware, allowing operators to mix and match best‑of‑breed components while maintaining a consistent networking experience.
Target Audience
Primary customers are AI cloud service providers and GPU‑focused data center operators that need to deliver secure, multi‑tenant networking at scale, as well as hyperscale CSPs integrating AI workloads into their existing infrastructure.
Features
- Unified automation for Ethernet, InfiniBand, NVLink, and DPU fabrics via a single control plane
- Auto‑generation of VRFs, VXLANs, BGP, ACLs, and routing policies with sub‑two‑minute provisioning
- Cloud‑grade network abstraction (VPC, elastic IP, load balancer, peering) mapped to hardware configurations
- Hard multi‑tenancy enforced at the switch/DPU level, supporting bare‑metal, VM, and container workloads
- RESTful API and Terraform provider for CI/CD integration and infrastructure‑as‑code workflows
- Closed‑loop assurance engine that validates topology, detects misconfigurations, and triggers auto‑remediation
- Compatibility with NVIDIA Spectrum and other leading switch hardware, with a hardware‑requirements matrix
- Role‑based access control and audit logging for compliance‑focused environments