July

Info

GPU PaaS releases are initially rolled out via Rafay's Air Gapped Controller form factor. These will be periodically bundled and rolled out into Rafay's Production SaaS.

v3.1-31¶

04 July, 2025 (est.)

Host Cluster Capacity Check for vClusters¶

The platform has been enhanced to keep track of the Host Cluster’s node-level capacity. So, when a vCluster creation request comes in, the platform will identify and allocate a node that can satisfy the requested GPU type and count.

Info

The end user will be presented with an error message if capacity for the vCluster cannot be allocated.

Branding & Theme Enhancements¶

This release builds on branding and theme enhancements from the previous release. These changes provide greater branding flexibility and a unified experience.

Admins can now configure custom primary, secondary, and sidebar colors, expanding beyond the previously limited primary/secondary options.

Instant Preview¶

Admins are provided with an instant preview showing what the updates to the theme would look like.

Reset to Default¶

Admins can quickly reset changes to default by clicking on the reset button.

Logos¶

In the Ops Console, admins can now provide the logos for dark and light background modes.

Favico & Background Images¶

In the Ops Console, admins can now provide the favico and background images for the web console.

URLs in Self Service Portal¶

In the Ops Console, admins can now specify custom URLs for the following.

Terms of Service
Privacy Policy, and
Online Documentation

These will then be dynamically updated in the end user facing web console in the login and sign up pages

End user login and account sign up pages now use the admin configured theme ensuring consistent themes across the entire platform.

Shown below is an example of a themed login page.

Shown below is an example of a themed sign-up page.

Schedules¶

Schedules allows for automated start/stop/custom actions on resources based on cron expressions and time zone. Schedules can be defined at various levels:

Project Tags
Compute/Service Profile Spec
Instance / Service Spec

Example 1

A good example for a use case for this is "Time-Sharing Limited GPU Resources Across Teams". For example, consider an enterprise that has a limited pool of high-end GPUs (e.g., A100s or H100s) that must be shared among multiple internal teams (e.g., research, inference, and training teams). To prevent resource contention and ensure fair access, teams are assigned non-overlapping usage windows.

How Schedules Help:

Research team: 8:00 AM – 12:00 PM IST
Training team: 12:30 PM – 4:30 PM IST
Inference team: 5:00 PM – 9:00 PM IST

Benefits:

Enables fair and scheduled GPU sharing
Prevents one team from monopolizing resources
Automates transitions and avoids manual intervention
Scales easily across projects using profile-level or tag-based schedules

Example 2

Shift-Left Provisioning for Slow-Starting Workloads

Some environments—such as GPU-backed training clusters, large LLM inference stacks, or complex data pipelines—can take 10–30 minutes to provision and become operational. Users expect these environments to be ready when they log in, not wait for deployment delays. This is particularly impactful in education, enterprise R&D, or MLOps platforms, where slow-starting resources can derail daily workflows or testing cycles.

Use a “start” schedule to pre-warm or provision environments in advance of expected usage windows. For example,

Environment scheduled to start at 7:30 AM IST
Users typically log in by 8:00 AM IST
Auto-stop schedules to shut down idle environments post-work hours.

Benefits:

Eliminates cold-start delays for time-sensitive teams
Improves user experience for researchers, analysts, or students
Ensures readiness without manual intervention
Optimizes resource scheduling across dependent systems (e.g., inference + storage + UI)

Email Notifications¶

Cloud Provider administrators can now configure email providers for user notifications. SMTP and SendGrid are the currently supported providers. When enabled, end users will automatically receive notifications for scenarios such as the following:

Temporary password for new users
Password reset workflows
Self service sign up of new orgs, etc

v3.1-32¶

18 July, 2025 (est.)

Container Deployment Wizard¶

End users can use Rafay's wizard type workflow to deploy their container images to the Kubernetes cluster or virtual cluster based compute instances without having to write a single line of Kubernetes YAML. Based on the user's inputs, Rafay will automatically generate a deployment YAML specification and deploy it to an isolated namespace in the shared/tenant specific cluster.

Pod as a Service¶

A new type of compute (Pod-as-a-Service) is now available for providers to offer their end users.

Users can request environments via CLI or via Rafay's self service portal by specifying "OS, container image, GPU, and SSH keys". Unlike a VM or a Kubernetes Cluster, the environment is provisioned in ~30 seconds. The end user is then provided with seamless SSH and HTTPS web access to the deployed environments.