Opportunities

NVIDIA® DOCA™ unlocks the potential of the NVIDIA BlueField® data processing unit (DPU) to offload, accelerate and isolate workloads in data centers. With DOCA, developers can program the data center infrastructure of the future by creating software-defined, cloud-based, DPU-accelerated services with zero-trust protection to meet the growing performance and security demands of today’s data centers.

At the core of this innovation is DOCA Flow, an API paramount to building agile execution pipelines in hardware. The library provides developers with a powerful toolkit for building sets of pipes, each characterized by compliance criteria, monitoring capabilities, and a set of actions. These pipes can be interconnected, allowing packages to seamlessly transition from one pipe to another after executing pipe-defined actions.

Enroll in this course to embark on a journey of DOCA Flow programming personified by developing an “ARP Storm Control” application. This application prevents network outages resulting from broadcast storms by skillfully forming a DOCA Flow pipeline that quells malicious broadcast network activity while preserving the integrity of well-managed traffic.

Key Learning Objectives:

• Understand the key role that DOCA Flow occupies in the broader DOCA and DPU landscape.
• Create an offloaded virtual network function (VNF) application.
• Acquire the skills to accelerate network applications through the DPU’s embedded switch.

Prerequisites:

• Basic understanding of networking fundamentals to optimize your participation in the course.

Introducing the NVIDIA License System (NLS), a new licensing solution designed to support the expansion of NVIDIA’s range of enterprise software. Designed to outperform NVIDIA’s existing software license server for virtual GPUs (vGPUs), NLS features advanced functionality and offers a number of licensing service use cases – Cloud License Service (CLS) and Delegated License Service (DLS).

During this course, attendees will enter the realm of NLS and gain a comprehensive understanding of the transition from their current licensing system to NLS. The course covers the following key aspects:

1. How to install, configure and use a Cloud License Service (CLS) and Delegated License Service (DLS) instances.
2. Migrate your existing NVIDIA Virtual GPU (vGPU) Software License Server to NVIDIA License System (NLS).

The course topics are:

• Introduction to the NVIDIA License Service
• Installation and Configuration of the Cloud License Service (CLS)
• Installation and Configuration of the Delegated License Service  (DLS)

The course objectives are:

• Оverviewing the NVIDIA License System (NLS)
• Discussing the benefits of the NVIDIA License System (NLS)
• Understanding the two instance types: Cloud License Service (CLS) and Delegated License Service (DLS)
• Explaining the steps required to install and configure each of the two instance types
• Walkthrough migration steps from the Legacy vGPU License System to the NVIDIA License System (NLS)

The prerequisites are:

• Basic knowledge of installing and setting up virtual appliances.
• Familiarity with web/cloud applications.
• Familiarity with NVIDIA products such as vGPU and NVAIE.

Explore the NVIDIA Omniverse development platform for virtual world builders and creators, tailored for virtual realm virtuosos. Enter the complex realm of UI mastery with an in-depth exploration of the robust omni.ui toolset and framework within the NVIDIA Omniverse Kit. In this self-paced course, hands-on exercises will guide you to create a custom UI that streamlines Omniverse workflows.

During this workshop, participants will gain the skills to:

• Initiate Omniverse Code with ease.
• Manage extensions by installing, uninstalling and activating them in Omniverse Code.
• Navigate the extensions panel, opening views and selecting primitives.
• Seamlessly navigate USD’s complex hierarchy of stages.
• Construct a concise form window, complete with stylistic elements.
• Test the limits of one’s creativity 
• Accept the challenge to mimic a window design created by the Omniverse UX team.
• Match form and function to the defined window design.

Once you’ve achieved the culmination, you’ll not only have created a user interface, but you’ll also have gained an in-depth understanding of the Omniverse code.

Prerequisites:

• Basic knowledge of Python (helpful, but not required).
• The Python Tutorial.
• Tools used:
• Omniverse Code
• Visual Studio Code
• Python
• Python Extensions

Minimum recommended hardware:

• Desktop or laptop computer equipped with Intel i7 Gen 5 or AMD Ryzen.
• NVIDIA RTX-capable GPU with 16 GB of capacity.

Similar courses:

“Getting Started with USD for Collaborative 3D Workflows,” a dive path into the Universal Scene Description (USD) framework developed by Pixar designed to foster collaboration between the creative and technical creators of the world.

Embark on an interactive journey through the “Assembling a Simple Robot” tutorial as you dive into creating a dynamic mobile robot on two wheels in a live Isaac Sim GPU environment. The immersive tutorial experience lasts approximately 30 minutes.

Within this lab you will master the art of:

• Establishing a connection between a local streaming client and an Omniverse Isaac Sim server located in the cloud.
• Introducing a mock USD robot into the live Isaac Sim environment.
• Inserting the robot body with joint actuators and properties.
• Enhancing the robot’s capabilities by integrating articulations.
• Upon completion of the course, you will gain a basic understanding of the Isaac Sim interface and core documentation, which will help you embark on your own robot simulation endeavors.

Prerequisites:

• Windows or Linux based computer equipped to install Omniverse Launcher and Omniverse applications.
• Internet bandwidth that can accommodate the Isaac Sim client/server stream (note that performance may vary).

Tools:

• NVIDIA Omniverse™ Isaac Sim

Additional Training:

For further exploration, consider enrolling in the Getting Started with USD for Collaborative 3D Workflows course, a self-paced adventure that delves into creating scenes using Universal Scene Description ASCII (.USDA) files that are human-readable.

Want to customize the Omniverse experience to your preferences? Dive into the realm of customization with Python-coded extensions. These versatile tools give you the ability to change everything from the functionality to the UI of Omniverse. Whether you seek to effortlessly create objects, introduce custom physics for select objects, streamline repetitive operations, or revolutionize object manipulation in the UI, extensions pave the way.

Discover the art of:

• Creating a button to spontaneously create a cube.
• Creating custom windows to change the Omniverse user interface.
• Debugging and publishing extensions.

Once complete, you’ll be able to design and share custom extensions in the thriving Omniverse community.

Basic knowledge of:

• Python, establishing a solid foundation.
• Computer graphics basics, but not required.

Recommended tools:

• Omniverse Code
• Visual Studio Code
• Python

Hardware requirements (minimum recommendations):

• Linux or Windows operating system.
• (RTX is optional for Omniverse Code.)
• Intel I7 Gen 7+ or AMD Ryzen+.

For further exploration, check out these relevant trainings:

• “Getting Started with USD for Collaborative 3D Workflows”
• “Easily develop advanced 3D layout tools in NVIDIA Omniverse”
• “Build beautiful, custom UI for 3D tools in NVIDIA Omniverse”
• “How to Build Custom 3D Scene Manipulator Tools on NVIDIA Omniverse”

NVIDIA has created a set of comprehensive reference applications including Create, Drive Sim and Isaac Sim to demonstrate the capabilities of the Omniverse Kit. Everything you see in these reference apps extends to your custom apps. You can leverage existing NVIDIA extensions to speed up your application development, as applications consist of different extensions that collaborate to streamline specific workflows. 

During this course, participants will understand:

• The essence of the Kit file and the art of its creation.
• The process of integrating extensions into applications.
• The role of extension dependencies.
• How to formulate the application layout.
• The use of styling techniques.
• Packaging and distributing an application.

Once completed, you will be able to create and share your own custom applications within the dynamic Omniverse community.

Prerequisites:

• Basic knowledge of Python.
• Elementary knowledge of computer graphics may be helpful, but is not required.
• Developing extensions for Omniverse.
• GitHub navigation.
• Command line proficiency.
• Tools in action:
• Visual Studio Code
• Python

Minimum recommended hardware:

• Desktop or laptop computer with Intel i7 Gen 5 or AMD Ryzen.
• GPU with NVIDIA RTX support and 16 GB.
• Connected Learning:
• “Getting Started with USD for Collaborative 3D Workflows”
• “Easily develop advanced 3D layout tools in NVIDIA Omniverse”
• “Building a beautiful, custom user interface for 3D tools in NVIDIA Omniverse”
• “How to create custom tools for manipulating 3D scenes in NVIDIA Omniverse”

Related Training:

• Getting Started with USD for Collaborative 3D Workflows
• Easily Develop Advanced 3D Layout Tools on NVIDIA Omniverse
• Build Beautiful, Custom UI for 3D Tools on NVIDIA Omniverse
• How to Build Custom 3D Scene Manipulator Tools on NVIDIA Omniverse

From recognizing a cat in the backyard to enhancing the customer shopping experience, the possibilities of AI-based video understanding know no bounds. Enter the NVIDIA Jetson Nano Developer Kit, a user-friendly yet powerful computer capable of running multiple neural networks simultaneously. Positioned as an ideal entry point into the field of intelligent video analysis (IVA) through the NVIDIA DeepStream SDK, this course gets you into the realm of building insightful projects. With JupyterLab notebooks and sample Python applications on your Jetson Nano, you’ll explore the field of deep learning video analytics by discovering deep insights from video streams. The skills you’ll develop will pave the way for creating your own future projects not only on the Nano, but on all Jetson Edge platforms.

Prerequisites:

• Basic familiarity with the Linux command line.
• Mastery of basic Python 3 programming concepts including functions, loops, dictionaries, and arrays.
• Tools in Action:
• DeepStream
• TensorRT
• Jetson Nano
• Python

Basic Hardware:

• NVIDIA Jetson Nano Developer Kit or NVIDIA Jetson Nano 2GB Developer Kit.
• Depending on the model, recommended power supply and jumper configurations.
• High performance microSD card (minimum 64 GB).
• USB webcam (for final use on the notebook).

Additional requirements:

• For the Jetson Nano 2GB developer kit, USB-C 5.1V 3A power supply from the list of supported components
• High performance microSD card: 64 GB minimum recommended (we have tested and recommend this one)
• USB data cable: Micro-B to Type-A with DATA function enabled (we’ve tested and recommend this one)
• USB Webcam (used only in the end notebook), such as Logitech C270 USB Webcam

Additional computer and network requirements:

• A computer with an internet connection and the ability to flash your microSD card if you are setting up the Jetson Nano for the first time
• Available USBA port on your computer (you may need an adapter if you only have USBC ports)
• Administrative rights and the ability to install compatible media player software (we tested and recommend this one)
• Internet connection for your Jetson Nano

Connected Learning:

• “Getting Started with Jetson Nano Artificial Intelligence“
• “Getting Started with Artificial Intelligence on Jetson Nano“

There is a certificate upon the completion of the course.

Accompanying Mark Harris’s renowned blog post, “An Even Easier Introduction to CUDA,” this notebook guides you through the fundamentals of crafting CUDA kernels for NVIDIA GPUs in a massively parallel manner. By completing this notebook, you’ll gain the ability to:

– Launch CUDA kernels massively in parallel on NVIDIA GPUs.

– Organize parallel thread execution for extensive dataset sizes.

– Manage memory interchange between the CPU and GPU.

– Profile your CUDA code for performance enhancements.

Upon finishing, you’ll be well-versed in launching massively parallel CUDA kernels for processing data on NVIDIA GPUs.

Prerequisites:

While no novel coding is required, a familiarity with the following will enhance your comprehension of the material:

– Writing, compiling, and running C or C++ code.

Recommended resources for prerequisites:

– The interactive tutorial at learn-c.org.

Tools utilized:

– CUDA C++

– nvcc

– nvprof

Related Training:

– “Fundamentals of Accelerated Computing with CUDA C/C++“: A self-paced course tailored to novice CUDA C++ programmers.

– “Fundamentals of Accelerated Computing with CUDA Python“: A self-paced course designed for those new to CUDA Python programming.