Search Our Database

Introduction

Deploying a graphics processing unit (GPU) instance using a standard operating system image within the IP ServerOne Customer Portal is a vital process for handling high-performance computing workloads, machine learning tasks, and graphical rendering. This capability allows administrators and developers to provision powerful virtual servers equipped with dedicated hardware acceleration on the Novagpu platform. By utilizing standard OS images optimized for GPU integration, users can quickly set up a reliable and robust environment for advanced computational projects. This guide provides step-by-step instructions on how to successfully provision and launch a new GPU cloud instance through the portal interface.

Prerequisite

• Active account access to the IPServerOne Customer Portal with appropriate permissions
• Sufficient GPU resource quotas and compute allocations within your Novagpu project.
• Sufficient account cloud credit

Before diving into the launching cloud instance, you may refer to the guidance below, as there are being used across this guidance.

• To register and login into our Customer Portal
  -> [How to register and login to Customer Portal]
• To configure Security Group for the cloud instance, you may consider as the simple firewall for the cloud instance
  -> [Configure Security Group in IPServerOne Customer Portal]
• If you wish to unify the Public Key for all cloud instance, you may generate a Public Key before launching the cloud instance
  -> [How to Create Public Key for the Cloud Server in Customer Portal]
• If you have existing Public Key, you may consider import the Public Key
  -> [How to Import Public Key(s) for the Cloud Server in Customer Portal]
• To create a Project for your cloud instance for easier managing, and you are having an option to assign the Project later
  -> [Managing project and instance assignment in Customer Portal]
• To login the cloud instance via SSH
  -> [How to login to your Linux Server via SSH]
• To retrieve the Windows Administrator password for RDP access
  -> [How to Retrieve the Windows Administrator Password]

Step-by-Step Guide

Step 1: Access the Customer Portal

• Log in to the [IPServerOne Customer Portal].

Step 2: Navigate to Novagpu

• Go to the NovaCloud and select NovaGPU Instances to view your GPU instance management page.
  

Step 3: Initiate Instance Launch

• Click the Launch NovaGPU button to create a new GPU instance.
  

Step 4: Select Standard OS Image

• Under the Select instance type, choose Create Instance from Image as the type.
  

• Select your desired standard operating system image from the available list.
  

Step 5: Configure Resources and Network

• Configure your desired Region, Plan, Project, then enter the desired Storage Size and Instance Name.
  

• Select your Public Key and using Security Group to assign which ports to be open.
  

• In the pop-out windows, verify all the configured settings and resource allocations in the summary panel then click Create to initiate the instance creation process.
  

Step 6: Complete Deployment

• Navigate back to your GPU instance management page and you may see the new created GPU Instance.
  

Conclusion

By following this guidance, you can successfully launch and deploy a high-performance GPU instance using a standard OS image on the NovaGPU platform via the Customer Portal, enabling you to accelerate your compute-intensive applications and manage your infrastructure effectively.

If you encounter any issues or need further assistance, please contact our support team at support@ipserverone.com.

Article updated on 20 October 2026 by Johnny