Overview

aarna.ml’s GPU Cloud Management Software (CMS) enables GPU-as-a-service providers to build hyperscaler-grade AI clouds, empowering a transition from static GPU rentals to fully dynamic, multi-tenant, and on-demand infrastructure.

From an infra management perspective, aarna.ml GPU CMS supports CPU+GPU isolation, Ethernet and InfiniBand Network isolation, External Storage provisioning for tenant including storage network isolation, external network isolation, dynamic IaaS including BMaaS and VMaaS.

From an PaaS perspective, aarna.ml GPU CMS supports Kubernetes platform management and serverless job submission and model serving capabilities.

aarna.ml GPU CMS Infra Manager (Super Admin): Sr. No. Feature
1 Discover compute nodes and switches
2 Validate discovered infra against the original intent/planned deployment
3 Initiate automated underlay configuration
4 Define OS flavors and populate them in the catalog
5 Create Tenant Enterprise accounts
6 Create / Invite Tenant admin user accounts
7 Observe and monitor the entire AI cloud infra
8 Define RBAC for tenant account
9 Define quotas for tenant account
10 Specify billing parameters
11 Set FinOps alerts, enable / disable access to enterprise users
aarna.ml GPU CMS Infra Manager (Tenant Admin): Sr. No. Feature
1 Create end users under tenant
2 Define RBAC for end user accounts
3 Define quotas for end user account
4 Provision BM instances with desired OS flavor (Optional)
5 Create VMs with desired flavors for tenant org users (Optional)
6 Observability and monitoring of the AI cloud infra allocated to enterprise org
7 Set FinOps alerts, enable / disable access to end users
8 Billing interface for the Tenant
aarna.ml GPU CMS Infra Manager (Tenant User): Sr. No. Feature
1 Provision BM instances with desired OS flavor
2 Create VMs with desired flavors, and manage their life cycle
3 Create, manage (mount, unmount, delete etc.) storage instances
4 Observability and monitoring of end user resources
aarna.ml GPU CMS PaaS (Tenant Admin): Sr. No. Feature
1 Create and manage clusters (add / remove nodes) comprising BM / virtualized compute nodes
2 Create and manage Kubernetes clusters with multiple master and worker nodes (add / remove nodes)
3 Define cluster wide policies
4 Create / Invite end users
5 Provision external storage for the clusters
6 Define policies and quota for end users
7 Observe and monitor the clusters
8 Specify billing parameters
9 Set FinOps alerts, enable / disable access to enterprise users
aarna.ml GPU CMS PaaS (Tenant User): Sr. No. Feature
1 Access PaaS resources
2 Submit HPC, AI/ML custom jobs / workloads
3 Monitor job status / workloads
aarna.ml GPU CMS Self-Service: Sr. No. Feature
1 Maintain LLM catalog (add, update, delete)
2 Manage integrations with model repositories (add, update, delete)
3 Specify billing parameters
4 Create / Invite end users
5 Observe and monitor all end user resources
6 Set FinOps alerts, enable / disable access to end users
7 Select and consume AI/ML service
8 Submit / view job outputs
9 Create explicit GPU resources
10 Monitor usage
Note: Sr. No. 1-6 typically are Tenant Admin features and Sr.No. 8-10 are Tenant User Features

Resources

aarna.ml GPU Cloud Management Software (CMS) Demo Detailed demo of the various capabilities of aarna.ml GPU Cloud Management Software
On-demand Multi Tenancy for GPUaaS Providers Documentation on how NVIDIA Cloud Partners and GPUaaS providers can add on-demand multi tenancy
aarna.ml GPU Cloud Management Software (CMS) Page Overview of the aarna.ml GPU Cloud Management Software (CMS) product

Certifications

Learn about Red Hat Certification and Partner Validation

Red Hat certified products are tested to meet Red Hat’s criteria and supported as defined in the Red Hat Collaborative Support Process.

Partner validated products are tested by Red Hat Partners and supported as defined in the Red Hat Third Party Component Policy.

Certification standards

This product hasn't been tested by Red Hat

Deploy & use

Quick start and configuration

Getting started

FAQs

What are the system requirements for installing Aarna.ml CMS?

Operating System: Ubuntu 20.04+
• CPU: Minimum 32 cores (64 recommended for production)
• Memory: Minimum 64 GB RAM
• Storage: At least 500 GB free disk space
• GPU (optional): For GPU-based workloads, ensure NVIDIA drivers and container runtime are installed
• Network: Internet access for pulling container images, DNS resolution

How is Aarna.ml CMS deployed?

Aarna.ml CMS is deployed as a set of containers using Kubernetes. It supports:
- Custom Ansible plug-in based installation
- On-premises and edge deployments
- Cloud-native environments

Does Aarna.ml CMS require a license or activation key?

- A license is required for production environments.
- For POCs, a trial license or credit may be provided (e.g., $50 via https://cloudbroker.aarna.ml).
- Licensing is tied to your organization registration.

How do I access the Aarna.ml CMS UI?

After setup, the CMS UI is available at:
- https://<your-domain-or-ip>/app
- Default login credentials will be shared during installation or can be created during setup.

Certifications

Learn about Red Hat Certification and Partner Validation

Certification standards

This product hasn't been tested by Red Hat

Quick start and configuration

Getting started

aarna.ml GPU Cloud Management Software (CMS) Partner Validated

Certifications

Quick start and configuration

Get started

Platforms

Products & services

Try, buy, sell

Help

About Red Hat Ecosystem Catalog

Red Hat legal and privacy links

Red Hat legal and privacy links

aarna.ml GPU Cloud Management Software (CMS) Partner Validated

Certifications

Quick start and configuration

Get started

Sr. No.	Feature
1	Discover compute nodes and switches
2	Validate discovered infra against the original intent/planned deployment
3	Initiate automated underlay configuration
4	Define OS flavors and populate them in the catalog
5	Create Tenant Enterprise accounts
6	Create / Invite Tenant admin user accounts
7	Observe and monitor the entire AI cloud infra
8	Define RBAC for tenant account
9	Define quotas for tenant account
10	Specify billing parameters
11	Set FinOps alerts, enable / disable access to enterprise users

Sr. No.	Feature
1	Create end users under tenant
2	Define RBAC for end user accounts
3	Define quotas for end user account
4	Provision BM instances with desired OS flavor (Optional)
5	Create VMs with desired flavors for tenant org users (Optional)
6	Observability and monitoring of the AI cloud infra allocated to enterprise org
7	Set FinOps alerts, enable / disable access to end users
8	Billing interface for the Tenant

Sr. No.	Feature
1	Create and manage clusters (add / remove nodes) comprising BM / virtualized compute nodes
2	Create and manage Kubernetes clusters with multiple master and worker nodes (add / remove nodes)
3	Define cluster wide policies
4	Create / Invite end users
5	Provision external storage for the clusters
6	Define policies and quota for end users
7	Observe and monitor the clusters
8	Specify billing parameters
9	Set FinOps alerts, enable / disable access to enterprise users

Sr. No.	Feature
1	Access PaaS resources
2	Submit HPC, AI/ML custom jobs / workloads
3	Monitor job status / workloads

Sr. No.	Feature
1	Maintain LLM catalog (add, update, delete)
2	Manage integrations with model repositories (add, update, delete)
3	Specify billing parameters
4	Create / Invite end users
5	Observe and monitor all end user resources
6	Set FinOps alerts, enable / disable access to end users
7	Select and consume AI/ML service
8	Submit / view job outputs
9	Create explicit GPU resources
10	Monitor usage