A Scheduling Framework for Decomposable Kernels on Energy Harvesting IoT Edge Nodes
Open Access
Author:
Jose, Sethu
Graduate Program:
Computer Science and Engineering
Degree:
Master of Science
Document Type:
Master Thesis
Date of Defense:
March 16, 2022
Committee Members:
Mahmut Taylan Kandemir, Thesis Advisor/Co-Advisor John Morgan Sampson, Committee Member Chitaranjan Das, Program Head/Chair Vijaykrishnan Narayanan, Committee Member
Keywords:
Internet of Things Non-Volatile Processors Energy Harvesting Approximate Computing Scheduling. Internet Of Things Scheduling
Abstract:
With the growing popularity of the Internet of Things (IoTs), emerging applications demand that edge nodes provide higher computational capabilities and long operation times while requiring minimal maintenance. Ambient energy harvesting is a promising alternative to batteries, but only if the hardware and software are optimized for the intermittent nature of the power source. At the same time, many compute tasks in IoT workloads involve executing decomposable kernels that may have application-dependent accuracy requirements. In this work, we introduce a hardware-software optimization framework for such kernels that aims to achieve maximum forward progress while running on energy harvesting Non-Volatile Processors (NVP). Using this framework, we develop an FFT and a convolution accelerator that computes 3.7x faster, while consuming 5.4x less energy, compared to a baseline energy-harvesting system. With our accuracy-aware scheduling strategy, approximate computing in this framework delivers on average 6.6x energy reduction and 4x speedup by sacrificing minimal average accuracy of 6.9%
