Project Description
Electronic waste, often referred to as e-waste, is a growing environmental crisis. Each year, the world disposes of a staggering 50 million tons of e-waste, of which more than 85% is buried in landfills. This not only consumes valuable land resources but also releases hazardous heavy metals into our environment. Such a concerning scenario poses risks not just to our surroundings, but also directly impacts human health. As computer scientists and system designers, we possess the potential to change this narrative. The designs we adopt today, and the components we choose to use, can dramatically influence the amount of e-waste generated tomorrow. However, to make informed decisions that prioritize sustainability, we must first quantify the environmental footprint of our design choices. This project will help to develop models of e-waste that will be useful in the design of new computer systems and will be a powerful tool for computer architects looking to design new, more efficient, and more environmentally friendly computer systems.
The heart of this project lies in accurately determining the materials and components present in electronic systems. Specifically, how do the numerous tiny components on server-class circuit boards contribute to the e-waste conundrum? Before we can design sustainably, we must understand what we're working with, down to the smallest capacitor. Our team will gather images and data about a variety of server class machines, concentrating on the circuit boards, and will take pictures of each of them and then use a variety of sampling and image processing techniques to create a data set that can then be used for machine learning estimates of component types. Are you passionate about sustainability? Are you interested in computer systems engineering? Do you want to be part of the solution to one of the 21st century's most pressing technology challenges? This is your chance to make a difference and help pave the way for a sustainable computing future.
Prerequisite Information
Introductory programming, some exposure to chemistry preferred.
Knowledge/Skills to Acquire (with guidance from mentors)
- What goes into a modern printed circuit board
- How heavy metals impact both the environment and human health
- Methods for statistical estimation useful in a variety of data science contexts
- Introductory image processing and machine vision techniques
Team Members
- Samantha West
- Claire Pemberton
- Ivan Hernandez
- Mariana Rosillo
Professor and Mentors
- Prof. Timothy Sherwood
- Grad mentor: Pranjali Jain
Meeting Times
- Mentor Meetings
- Fridays, 12-1 p.m.
- ERSP Team Meetings
- Fridays, 1-3 p.m.
Research Logs
References
[1] Udit Gupta, Young Geun Kim, Sylvia Lee, Jordan Tse, Hsien-Hsin S. Lee, Gu-Yeon Wei, David Brooks, Carole-Jean Wu, Chasing Carbon: The Elusive Environmental Footprint of Computing