Automate the Infrastructure Behind AI Factories

Connect Kubernetes provisioning with AI network orchestration. Clastix and Aviz help GPU cloud providers, neoclouds, and enterprises deploy multi-tenant GPU environments with coordinated compute and network operations.

Skip the browsing. Ask AI.

Hello! How may I help you today?

Why this partnership matters

Bring Compute and Network Provisioning Into One Workflow

AI infrastructure requires more than GPU servers. Operators must provision Kubernetes environments and connect compute to the network, storage, and services workloads depend on. Clastix and Aviz bring these operational layers closer together.

Bring Environments Online Faster

Coordinate Kubernetes and GPU infrastructure provisioning with automated network configuration.

Extend Multi-Tenancy

Carry tenant requirements from the compute environment into the network fabric for isolated connectivity.

Automate Network Changes

Reduce repetitive switch-by-switch work as GPU servers, clusters, and tenant environments grow.

Improve Operational Consistency

Align infrastructure lifecycle activities with network provisioning and ongoing operations.

What the joint solution delivers

Kubernetes Infrastructure Meets AI Network Automation

Clastix manages Kubernetes environments across infrastructure, including bare metal. Aviz automates the network fabric that connects GPU compute, tenants, and services.

Clastix kMetal Platform

Kubernetes infrastructure management

Kubernetes Provisioning and ManagementProvision and manage Kubernetes environments across infrastructure, including bare-metal deployments.
kMetalClastix's platform for provisioning and managing Kubernetes directly on bare metal, with no separate hypervisor layer.
Cloud-Native Infrastructure OperationsSupport Kubernetes environment lifecycle and cloud-native workloads while retaining control of the underlying infrastructure.

Aviz AI Factory Portfolio

Network automation, connectivity & operations

ONES AINVIDIA-certified network automation to bootstrap, orchestrate, and enable multi-tenancy across AI network fabrics.
Aviz Service GatewaySoftware-defined tenant connectivity using BGP/EVPN, VRF/VXLAN, NAT, elastic IPs, and stateful L3/L4 policies.
AI ParserToken-level visibility into inference usage, latency, and throughput by tenant and application.
AI NOCAgentic AI operations with customizable SRE skills to support troubleshooting and operational workflows.

Where it works

Designed for Shared, High-Performance AI Infrastructure

Support environments where Kubernetes clusters, GPU compute, and network fabrics must work together.

AI Factories

AI Factories & GPU Clusters

Coordinate Kubernetes deployment and network automation for GPU infrastructure and AI fabrics.

Neoclouds

Neocloud Infrastructure

Simplify GPU environment deployment, tenant onboarding, and network service delivery.

Multi-Tenancy

Multi-Tenant GPU Environments

Extend tenant-aware provisioning into isolated network connectivity across shared GPU resources.

Bare Metal

Kubernetes on Bare Metal

Provision Kubernetes environments and automate their integration with the underlying network.

Enterprise AI

Enterprise AI Infrastructure

Establish repeatable compute and network operations for enterprise AI workloads and shared platforms.

Day-2

Ongoing Infrastructure Operations

Manage network changes as tenant environments evolve and compute capacity is added or removed.

Integrated infrastructure workflow

From Kubernetes Provisioning to Network Readiness

A coordinated operating model connects cluster lifecycle activities with the network configuration required by GPU workloads and tenant environments.

Kubernetes environments
Bare-metal provisioning
Cluster lifecycle

ONES AINetwork bootstrapping
Fabric orchestration
Multi-tenant networking

AI InfrastructureGPU servers and clusters
Tenant connectivity
AI workloads and services

NVIDIA GPU InfrastructureNVIDIA Spectrum-X EthernetNVIDIA InfiniBandNVIDIA BlueFieldMulti-vendor NetworkingBare-metal Kubernetes

Why Aviz with Clastix

Make Network Automation Part of the Kubernetes Infrastructure Journey

Provisioning a cluster is only one part of bringing an AI environment online. The network must also be configured to connect workloads, tenants, and shared services.

Coordinated Infrastructure Provisioning
Connect Kubernetes environment deployment with network configuration through Clastix and Aviz.
Tenant-Aware Network Orchestration
Extend tenant requirements into the AI fabric to support isolated connectivity across shared infrastructure.
Automation as Infrastructure Scales
Reduce manual network configuration as GPU servers, clusters, and tenant environments are added.
Day-0/1/2 Network Operations
Support initial deployment, service onboarding, and ongoing network lifecycle changes.
Open, Multi-Vendor Networking
Apply a consistent network automation approach across supported hardware and AI infrastructure architectures.