Designing with Versal™ AI Engine

(ref.ACAP_AIE)

4 days - 28 hours

Objectives

  • After completing this comprehensive training, you will have the necessary skills to:
    • Describe the Versal ACAP architecture at a high level
    • Describe the various engines in the Versal ACP device and the motivation behind the AI Engine
    • Describe the architecture of the AI Engine
    • Describe the memory access structure for the AI Engine
    • Describe the full application acceleration flow with the Vitis tool
    • Enumerate the toolchain for Versal AI Engine programming
    • Explain what intrinsic functions are
    • Program a single AI Engine kernel using the XChessDE tool
    • Program multiple AI Engine kernels using static data flow (SDF) graphs
    • Describe the system-level flow, which includes PS + PL + AIE (SW-HW-SW) designs
    • Describe the supported emulation for a system-level design
    • Describe the data movement between the PS, PL, and AI Engines
    • Describe the implementation of the AI Engine core and programmable logic
    • Implement a system-level design for Versal ACAPs with the Vitis tool flow
    • Utilize advanced MAC intrinsic syntax and application-specific intrinsics such as DDS and FFT
    • Utilize the AI Engine DSP library for faster development
    • Apply location constraints on kernels and buffers in the AI Engine array
    • Apply runtime parameters to modify application behavior
    • Debug a system-level design

Partners

xilinx atp

Prerequisites

  • Comfort with the C/C++ programming language
  • Software development flow
  • Vitis software for application acceleration development flow

Configurations

  • Software Configuration :
    • Vitis unified software platform 2020.2
  • Hardware configuration:
    • Recent computer (i5 or i7)
    • OS Linux 64-bits (Windows 10 compatible)
    • At least 16GB RAM
    • Display resolution recommended 1920x1080

Outline

Day 1

Overview of Versal ACAP Architecture

Introduction to the Versal AI Engine Architecture

Versal AI Engine Memory and Data Movement

Versal AI Engine Tool Flow

Application Partitioning on Versal ACAPs

Data Types: Scalar and Vector Data Types

Intrinsic Functions

Day 2

Window and Streaming Data APIs

The Programming Model: Single Kernel

The Programming Model: Introduction to the Data Flow Graph

The Programming Model: Multiple Kernels Using Graphs

Day 3

Application Partitioning on Versal ACAPs

ACAP Data Communications

System Design Flow

Introduction to Advanced Intrinsic Functions

Day 4

AI Engine DSP Library Overview

Advanced Graph Input Specifications

AI Engine Application Debug and Trace

Teaching Methods

  • Classroom training:
    • Face to face
    • Presentation by video projector
    • Provision of PDF course materials
  • Virtual training:
    • Onlive training
    • Presentation by Webex
    • Provision of PDF course materials

Support

  • Authorized Trainer Provider XILINX : Engineer Electronics and Telecommunications ENSIL
    • Expert FPGA XILINX - Language VHDL/Verilog - RTL Design
    • Expert SoC & MPSoC XILINX - Language C/C++ - System Design
    • Expert DSP & RFSoC XILINX – HLS - Matlab - Design DSP RF
    • Expert ACAP XILINX – AI Engines – Heteregenous System Architect

Methods of monitoring and assessment of results

  • Attendance sheet
  • Evaluation questionnaire
  • Evaluation sheet on:
    • Technical questionnaire
    • Result of the Practical Works
    • Validation of Objectives
  • Presentation of a certificate with assessment of prior learning

Concerned public

  • Software and hardware developers, system architects, and anyone who needs to accelerate their software applications using Xilinx devices