An Introduction to NVIDIA Omniverse Nucleus on AWS

Introduction

In this post we will discuss NVIDIA Omniverse, a platform for connecting and building custom 3D pipelines, and the potential it offers for Amazon Web Services (AWS) Spatial Computing customers. As well, we will provide high-level details about how AWS addresses important components of connecting Omniverse users.

Industry Trends & Challenges

In today’s environment, businesses are constantly innovating to stay competitive and relevant. Consumers have their own challenges they are trying to navigate. As a result, their behaviors can turn on a dime. Companies are working relentlessly to meet consumer expectations by reducing product cycle lifetimes.

The world has become more and more distributed, from global teams, to remote workers, increasingly intricate machinery, ever-growing datasets, and complex supply chains only becoming more complex. The focus on environmental sustainability means that companies are working to reduce the total carbon impact of products, from design to transport to material waste. From the boardroom to the shop floor, business leaders around the globe are continuously putting pressure on their organizations to innovate and address these challenges.

Spatial computing technologies are one area of innovation where businesses see opportunity to be more competitive and deliver value to the market. Underpinning spatial computing is 3D spatial data, a medium that is enabling organizations to tackle these challenges.

Innovating with AWS Spatial Computing

AWS Spatial Computing teams are innovating and solving industry challenges by making it easy to create, manage, operate, and deploy spatial computing experiences on AWS. By identifying pain points and developing building blocks, we are helping customers, like you, expand and simplify business operations by using 3D spatial data.

Spatial data is complex. There are a dizzying number of applications that can create, view, and update 3D data. There are equally many file interchange formats and conversion processes. In today’s demanding and fast-paced environments, the iteration of data must be quick. Your teams and stakeholders across all business lines need workflows that enable coordination and collaboration from their applications of choice. NVIDIA Omniverse on AWS solves these challenges.

What is Omniverse?

NVIDIA Omniverse is “a platform for building and operating metaverse applications, enabling enterprises of every size to develop custom 3D pipelines, and simulate large scale virtual worlds.”

The Omniverse platform is built on Universal Scene Description (USD), a powerful, extensible 3D framework and ecosystem that enables 3D designers and developers to connect and collaborate between industry-leading 3D content creation, rendering, and simulation applications. Omniverse helps individual creators to connect and enhance their 3D artistic process, and enterprises to build and simulate large scale virtual worlds for industrial applications.

With Omniverse, everyone involved in the lifecycle of 3D data has access to high-quality visualizations, authoring, and review tools. Teams do not need additional overhead to manage complex 3D data pipelines. Instead, they can focus on their unique contributions to bring value to the market. Non-technical stakeholders do not need to subject themselves to applications with steep learning curves, nor do results need to be compromised for the sake of iteration reviews.

What is Omniverse Nucleus?

A core feature of the Omniverse platform is Nucleus — the database and collaboration engine that enables real-time exchange of USD data. With Omniverse Nucleus, you can connect multiple users working in their preferred 3D applications, providing a single source of truth for their 3D data. This allows your teams to use the application they are the most comfortable with and opens the doors for rapid iteration.

NVIDIA Omniverse Nucleus operates under a publish-and-subscribe model and enables efficient live synchronization between Omniverse applications. It transmits changes to USD scenes in real-time between connected applications.

Omniverse Nucleus can be used by individual artists and creatives, deployed on a local workstation. Or, your enterprise teams can also deploy Omniverse Enterprise Nucleus Server on AWS.

Omniverse Enterprise Nucleus Server On AWS

To support distributed Omniverse users, Nucleus should be deployed in a secure environment. With on-demand compute, storage, and networking resources, AWS infrastructure is well suited to all spatial computing workloads, including Omniverse Nucleus.

With an extensive selection of compute instance types and storage solutions, you can right-size your infrastructure and fine-tune performance as needed. The AWS global footprint provides you the ability to connect distributed users from all over the globe. With the security, encryption, identity, and access management controls, you keep complete control over your data.

Omniverse running on AWS infrastructure opens a variety of use cases. In the retail industry, it is common for teams in different geographical locations to have ownership over each part of a product’s lifecycle. These teams need to coordinate around the same desired end goal while working in their tool of choice.

As another example, manufacturing companies sit on a treasure trove of Computer Aided Design (CAD) assets. These assets are well suited for manufacturing, but need to be converted to be consumed by visualization tools offered by Omniverse. With the AWS global storage solution, Amazon Simple Storage Service (Amazon S3), and content distribution networks, Amazon CloudFront, teams can easily deliver data to a Nucleus deployment and have it converted by Omniverse. Or, automate custom and third-party conversion processes with AWS Serverless technologies.

You can develop custom Omniverse Kit extensions that connect into AWS backend services to support event driven architectures.

With Omniverse RTX-powered visualization, simulation and out-of-the-box, life-like, render quality, Omniverse is an exceptional solution for synthetic data generation, data which can then be fed into
Amazon Machine Learning (Amazon ML) services for training models.

Next Steps

Explore Omniverse Enterprise today. Enterprises can purchase an Omniverse Enterprise Nucleus license, then learn how to Deploy Omniverse Enterprise Nucleus server on Amazon EC2.