Skip to content

kristopk/SDAccel-AWS-F1-Developer-Labs

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

SDAccel AWS F1 Developer Labs

Welcome to the SDAccel AWS F1 Developer Labs. By going through these labs you will gain hands-on experience with AWS F1 and learn how to develop accelerated applications using the Xilinx SDAccel development environment.

Prerequisites

In order to run the SDAccel AWS F1 Developer Labs, you need the following:

  • An AWS account
  • Access to AWS F1 instances
  • Familiarity with launching an AWS EC2 F1 instance with RDP services enabled

If you need help with any of the above, follow the setup instructions

Supported Versions

These labs are based on SDAccel 2019.1 available with the FPGA Developer AMI v1.7.0. Please make sure to select this version when you launch your instance. To find this AMI, search for "FPGA Developer AMI - 1.7.0" in the Community AMIs.

Developer Lab Modules

Introduction to the SDAccel Flow

  • Running the "Hello World" example
  • Introduction to the SDAccel development environment
  • Using the SDAccel GUI to optimize applications

FOLLOW THE INSTRUCTIONS

Optimize the data movement between CPU and FPGA

  • Profile the software application and determine which functions to accelerate on FPGA
  • Working with a predefined FPGA accelerator, optimize performance by overlapping data transfers and FPGA execution
  • Further speed-up the application by overlapping FPGA execution and CPU processing

FOLLOW THE INSTRUCTIONS

Methodology for Optimizing Accelerated FPGA Applications

  • Determine the software functions to be accelerated
  • Begin with a baseline design of 2D convolution of an RGBA video and a set of filter coefficients using ffmpeg
  • Perform a series of optimization methods to achieve the performance target

FOLLOW THE INSTRUCTIONS


Visit the SDAccel Developer Zone for more information about FPGA acceleration.

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • C++ 87.4%
  • C 10.6%
  • Makefile 2.0%