Technical paper titled “CFU Playground: Full-Stack Open-Source Framework for Tiny Machine Learning (tinyML) Acceleration on FPGAs,” from Google, Purdue University and Harvard University.
Abstract
“We present CFU Playground, a full-stack open-source framework that enables rapid and iterative design of machine learning (ML) accelerators for embedded ML systems. Our toolchain tightly integrates open-source software, RTL generators, and FPGA tools for synthesis, place, and route. This full-stack development framework gives engineers access to explore bespoke architectures that are customized and co-optimized for embedded ML. The rapid, deploy-profile-optimization feedback loop lets ML hardware and software developers achieve significant returns out of a relatively small investment in customization. Using CFU Playground’s design loop, we show substantial speedups (55x-75x) and design space exploration between the CPU and accelerator.”
Find the technical paper here. Published Jan. 2022.
arXiv:2201.01863v1 Shvetank Prakash, Tim Callahan, Joseph Bushagour, Colby Banbury, Alan V. Green, Pete Warden, Tim Ansell, Vijay Janapa Reddi.
Visit Semiconductor Engineering’s Technical Paper library here and discover many more chip industry academic papers.
100% inspection, more data, and traceability will reduce assembly defects plaguing automotive customer returns.
Engineers are finding ways to effectively thermally dissipate heat from complex modules.
Increased transistor density and utilization are creating memory performance issues.
Lots of unknowns will persist for decades across multiple market segments.
FPGAs, CPUs, and equipment receive funding in China; 98 startups raise over $2 billion.
Suppliers are investing new 300mm capacity, but it’s probably not enough. And despite burgeoning 200mm demand, only Okmetic and new players in China are adding capacity.
100% inspection, more data, and traceability will reduce assembly defects plaguing automotive customer returns.
From low resistance vias to buried power rails, it takes multiple strategies to usher in 2nm chips.
Some of the less common considerations for assessing the suitability of a system for high-performance workloads.
Manufacturing 3D structures will require atomic-level control of what’s removed and what stays on a wafer.
Different interconnect standards and packaging options being readied for mass chiplet adoption.
Engineers are finding ways to effectively thermally dissipate heat from complex modules.
Disaggregation and the wind-down of Moore’s Law have changed everything.
Leave a Reply