System Design

Designing PCIe Systems

Target Audience

Hardware designers, software engineers, and system architects who want to understand the deeper workings and create applications using the AMD Xilinx PCI Express solution

Description

This workshop provides a detailed explanation of how to implement an AMD Xilinx PCI Express® core in custom applications to improve time to market with the PCIe® core design.

The focus is on:

  • Constructing an AMD Xilinx PCI Express system within the customer education reference design
  • Enumerating various AMD Xilinx PCI Express core products
  • Leveraging key AMD Xilinx advantages related to performance, latency, and bandwidth in PCI Express applications
  • Identifying the advanced capabilities of the PCIe specification

Developing AI Inference Solutions

Target Audience

Software and hardware developers, AI/ML engineers, data scientists, and anyone who needs to accelerate their software applications using AMD Xilinx devices

Description

This workshop provides information on how to implement neural networks on cloud and edge platforms using the Vitis™ AI development platform.

The emphasis is on:

  • Illustrating the Vitis AI tool flow
  • Leveraging the architectural features of the Deep Learning Processor Unit (DPU)
  • Optimizing models using the AI quantizer and AI compiler
  • Utilizing the Vitis AI Library to optimize pre-processing and post-processing functions
  • Creating a custom platform and application
  • Deploying a design
  • Reviewing the benefits of the Kria™ K26 SOM board

Developing Multimedia Solutions

Target Audience

Engineers and designers who need to develop multimedia application targeting Zynq UltraScale+ MPSoC EV devices

Description

This workshop provides an explanation of how to build and run complex multimedia applications targeting Zynq® UltraScale+™ MPSoC EV devices with the help of the GStreamer framework. This course also illustrates how the use of the hardened video codec unit in the EV device helps to achieve optimum performance by offloading critical tasks to the dedicated processing engines.

The emphasis is on:

  • Describing the multimedia solutions provided by AMD Xilinx
  • Developing a multimedia application targeting Zynq UltraScale+ MPSoC EV devices
  • Listing the multimedia hardened blocks available in Zynq UltraScale+ MPSoC EV devices
  • Explaining the encoder and decoder functionalities of a video codec unit
  • Describing the software stack for developing multimedia applications
  • Describing the Vitis Video Analytics SDK (VVAS) technology
  • Utilizing the GStreamer framework from the software stack to create different multimedia pipeline

Vision Based Applications with Kria

Target Audience

Software and AI developers who want to get started with using Kria SOMs

Description

This workshop provides details about the Kria™ System-on-Module (SOM) and Kria KV260 Vision AI Starter Kit, enabling acceleration of vision-based applications using the KV260 Starter Kit without any prior FPGA knowledge.

The workshop also provides information on how to build hardware and software components, customize an AI model, and perform benchmarking. Design guidelines for developing a custom carrier card are also discussed.

The emphasis is on:

  • An overview of the operation and advantages of the Kria K26 SOM
  • An investigation of the Kria KV260 Vision AI Starter Kit
  • Running accelerated applications, such as the Smart Camera, AI Box, and Defect Detection applications, using the kit
  • Running the NLP SmartVision demo application with PetaLinux and Ubuntu images
  • Building hardware and software design components from scratch
  • Customizing the AI models used in the applications
  • Reviewing design guidelines for developers to design a custom carrier card