ASPLOS 2024

GeneSys Tutorial | April 28, 2024 | 1:30 PM - 5:00 PM PST | Grande D

Overview

This tutorial will present a discussion on emerging deep learning models along with a comprehensive walkthrough of our newly developed system, GeneSys. Attendees will gain valuable insights into our cutting-edge technology and its wide-ranging applications in the field of DNN acceleration systems.

Goals

We aim for attendees to gain a thorough understanding of the functionalities of our innovative system. They will learn how it can be effectively implemented in various applications and be equipped with the knowledge to harness its potential for their own projects and research in deep learning and DNN acceleration systems.

Who Should Attend

Researchers and developers interested in deep learning systems, compiler development, and hardware/software design for DNN acceleration systems.

Logistics

  • Venue: ASPLOS 2024
  • Date: April 28, 2024
  • Time: 1:30 PM - 5:00 PM PST
  • Registration: To secure your spot for our tutorial, please sign up for at least a “One-day Workshops/Tutorials Registration”. You can register here.

Schedule

Start (EST) End (EST) Agenda Presenter Resources
1:30 PM 1:50 PM Introduction Hadi slides
    History and challenges of hardware acceleration
Introduction to GeneSys
  video
1:50 PM 2:00 PM Motivation Rohan slides
    Neural networks and hardware acceleration
Inference pipeline – datacenters and edge
Systems challenges and opportunities
Overview of GeneSys
Example usage of GeneSys in research projects
  video
2:00 PM 3:00 PM GeneSys Architecture Soroush slides
    GeneSys NPU overview
Systolic array
On-chip memory architecture and memory interface
Tandem processor
End-to-end neural network execution
ISA
  video
3:00 PM 3:30 PM Afternoon Break    
3:30 PM 4:30 PM GeneSys Compiler Chris slides
    Introduction to compilation
Compilation challenges for DNN accelerators
f-DFG frontend
Codelet backend
Compiler overview
Using the compiler and it to your needs
Realities about end-to-end applications
FhY: Cross-domain compilation stack for multi-acceleration
  video
    Demo: Compiling ResNet50, changing tiling, loop order, on-chip buffers, fusing layers   video coming soon…
4:30 PM 5:00 PM GeneSys Uses, Verification, and Software Simulator Hanyang slides
    Example problem: data motion acceleration
Example problem: neuromodulation for brain-implantable devices
Quantization on GeneSys
Python APIs
Software simulator
RTL simulation
Validation with FPGA
  video

Presenters

  • Hadi Esmaeilzadeh: Halicioğlu Chair in computer architecture and professor at the University of California, San Diego
  • Soroush Ghodrati: System Architect at Apple
  • Rohan Mahapatra: Ph.D. student in computer science and engineering at the University of California, San Diego
  • Hanyang Xu: Ph.D. student in computer science and engineering at the University of California, San Diego
  • Christopher Priebe: Ph.D. student in computer science and engineering at the University of California, San Diego