Addressing Challenges of Cross-Architecture Dynamic Migration Over Heterogeneous Acceleration Systems
Offered By: Linux Foundation via YouTube
Course Description
Overview
Explore cross-architecture dynamic migration challenges in heterogeneous acceleration systems through this informative conference talk. Learn about China Mobile's innovative practices for enabling smooth AI application migration across diverse accelerators like GPUs, FPGAs, and NPUs. Discover solutions to tool chain and vendor lock-in issues, including a unified abstraction for accelerators, a middle-compiler approach for cross-architecture compilation, and a runtime supporting dynamic and replaceable linking. Gain insights into the architecture design, open source plans, and a practical demonstration showcasing how applications can migrate freely between accelerators without altering development habits.
Syllabus
Addressing Challenges of Cross-Architecture Dynamic Migration Over Heterogeneous Acce... Yanjun Chen
Taught by
Linux Foundation
Tags
Related Courses
Embedded Systems DesignIndian Institute of Technology, Kharagpur via Swayam FPGA computing systems: Background knowledge and introductory materials
Politecnico di Milano via Coursera Future of Computing - on the Road to Quantum
openHPI Learning FPGA Development
LinkedIn Learning Learn VHDL and FPGA Development
Udemy