Recent NSF Awards in Design for Environmental Sustainability in Computing Are Advancing US National Priorities
October 24, 2023
The U.S. National Science Foundation (NSF) has awarded $13M through its Design for Environmental Sustainability in Computing (DESC) program to 25 research teams seeking to contribute to technological innovation and environmental sustainability in computer systems. These grants encompass a wide range of projects including reducing greenhouse gas emissions and optimizing water consumption in data centers, creating novel methods to extend the useful lifetime of computing systems amortizing their carbon costs over longer periods of time, developing eco-friendly computing materials and optimizing algorithms for better use of energy from renewable sources.
“The DESC program demonstrates our clear commitment to advancing US national priorities,” said Margaret Martonosi, assistant director for the Computer and Information Science and Engineering directorate at NSF. “By promoting research and development that integrates environmental considerations into computing we are addressing critical environmental challenges and contributing to a sustainable future. As we continue to navigate an era of rapid technological advancement and environmental concerns, programs like DESC play a pivotal role in shaping the future of our nation and our planet.”
Headlining these awards are three Type I projects classified as large investments with budgets reaching up to $2M. The following institutions have been selected to receive the most recent Type I DESC awards:
The George Washington University and The Ohio University.
Researchers in this project aim to reduce the environmental pollution caused by the manufacturing of electronic devices and computers. Researchers will work on designing computer parts that can be reused to extend their lifespan. In addition, they will mix electronics and light-based technology to create a more efficient and reliable system for future computers.
University of Pittsburgh and University of Notre Dame.
The focus of this research project is to reduce carbon emissions from chip manufacturing by reusing recently retired computer chips and making them last longer. By connecting these chips together in new ways, the researchers will create devices that will compete with newly fabricated chips. This will help dramatically reduce pollution for developing new products while minimizing hazardous computer waste in landfills, which can cause serious health problems for humans, while still advancing computing capabilities.
Cornell University, Georgia Tech and Harvard University.
Life-time aware design frameworks for sustainable edge devices. This research project aims to make environmentally friendly electronic devices like wearable watches, phones, and sensors. These devices are a big part of our daily lives and are responsible for a lot of data we use, yet they can also harm the environment with the impact from their manufacture and are often treated as disposable. The researchers will create a toolkit called Delphi to help design these devices in a way that last longer and create less electronic waste.
The DESC program also funded more than a dozen Type II projects with investments of up to $600,000. The following institutions have been selected to receive the most recent Type II DESC awards:
The University of Texas at Arlington and University of California, Riverside.
A User-Interactive Approach to Water Management for Sustainable Data Centers: From Water Efficiency. to Self-Sufficiency. This research project focuses on reducing the water usage of data centers, which are huge users of water, by creating a system that tracks water use and helps users make sustainable choices. The project also plans to use alternative water sources like rainwater.
The Pennsylvania State University and Stony Brook University.
Extending lifetimes of partially broken machines to repurpose e-waste. The goal of this project is to improve sustainability of computing hardware by extending the lifetime of computing equipment even when they start to partially break down. Therefore, the research team will investigate a new approach in which, instead of repairing or throwing away computers, the remaining working components are utilized as-is.
Oregon State University and University of Florida.
Software-Hardware Recycling and Repair Dataset Infrastructure (SHReDI) for Sustainable Computing. This project aims to tackle the growing problem of electronic waste (e-waste)—the fastest-growing type of trash globally— by developing wireless tags that will provide recyclers with information about what materials are in electronic devices and how to recycle them properly. The information from the tags will help track recycled e-waste back to the manufacturers, which can help create better recycling policies.
Texas Tech University.
Dynamic Optimization for Leaps in Sustainability-Centric Computing System Design. This project focuses on the environmental impact of computing systems, which can contribute to issues like greenhouse gas emissions and electronic waste. Researchers aim to make computing systems more sustainable by using dynamic optimization techniques that can improve the way they work in real time to reduce their negative effects on the environment.
University of Missouri and The University of Texas at Dallas.
Sustainability-aware Reliable and Reusable AI Hardware Design. The research team on this project aims to make energy efficient Deep Neural Network (DNN) accelerators, which are designed for compute-intensive applications like artificial intelligence and machine learning. Currently, DNN accelerators use a lot of power, making their use limited for energy-constrained devices. Therefore, researchers will work on efficient methods that enable DNN reuse and self-repair while also optimizing performance and sustainability metrics.
University of California, Riverside.
Enabling Carbon-Zero Colocation Data Centers via Agile and Coordinated Resource Management. This research project focuses on making large data centers, like those used by tech companies and the AI industry, more eco-friendly. These data centers use a lot of energy, and the goal is to reduce their impact on the environment by developing new ways to coordinate renewable energy use, optimize server operations, and adjust server power usage in a smart way.
University of California, Riverside.
Minimizing Carbon Footprint by Co-designing Data Centers with Sustainable Power Grids. Data centers contribute to global greenhouse gas emissions in part because of the difficulty of matching energy demands of systems with renewable energy production in the grid. Researchers in this project seek to improve the carbon emissions produced by data centers by developing a fine grain micro-grid electricity models of renewable energy availability and propose a novel carbon metric that helps them better measure and reduce their impact on the environment.
Arizona State University and University of Minnesota, Twin Cities.
Towards Reduce- and Reuse-based Design of VLSI Systems with Heterogeneous Integration. Researchers in this project aim to reduce carbon emissions through minimized computation required for designing computing components from scratch. The team will develop a method for reusing parts of the computer chips and making them more efficient so that they use less energy and produce fewer harmful emissions.
University of Pittsburgh.
Data-driven system-design for sustainable long-lasting distributed infrastructures. This research project aims to reduce carbon emissions by transforming the current carbon-based, electricity-dependent distributed computing infrastructure. While previous efforts have focused on making computer systems more energy-efficient, this project takes a broader approach. It recognizes that merely reducing energy consumption may not cut carbon emissions. Instead, it seeks to develop innovative strategies for computing systems that consider their entire lifespan and emissions.
Southern Illinois University.
This project aims to substantially improve the sustainability of modern edge data centers by mitigating the environmental impact of artificial intelligence. Researchers will focus on promoting the adoption of sustainable practices in the computing industry, boosting the efficiency of data processing, and driving the transition of digital infrastructure towards a greener and more sustainable future.
Northwestern University and University of Pittsburgh.
Building Future-proof Learning-Enabled Cyber-Physical Systems with Cross-Layer Extensible and Adaptive Design. Researchers in this project aim to extend the lifespan of electronic devices to reduce the need to manufacture new ones as well as the e-waste that is produced when users discard them. The project will focus on smart devices like self-driving cars, medical equipment, and robots, which often need to adapt to constantly changing physical environment and have limited resources and stringent timing requirements.
Colorado State University.
Sustainable Serverless Computing. Cloud datacenters are a major contributor to global carbon emissions, and serverless computing will exacerbate these pressures on the global environment. This project will explore how serverless computing—the abstraction of compute, memory, data storage, networking resources, etc. from physical hardware—performs in datacenters with a goal to minimize overall environmental impact. Researchers will work on improving the way serverless computers start up, store data, and adapt to their environment.
Funding for these recent DESC awards came from NSF’s Computer and Information Science and Engineering directorate. The National Discovery Cloud for Climate (NDC-C) program and NSF’s Office of Advanced Cyberinfrastructure provided matching funds and support for minority serving institutions.
The DESC program promotes research and development that integrates environmental considerations into the design and utilization of computing and information systems. The program recognizes the critical role that computing technologies play in addressing environmental challenges while also seeking to address the environmental footprint of computing itself. The DESC program focuses on projects that meaningfully address sustainability beyond energy efficiency such as reducing greenhouse gas emissions, saving water consumption, and avoiding e-waste.
By promoting research and development that embeds environmental considerations into computing technologies, DESC is helping the nation address pressing challenges, foster innovation, and create a sustainable and competitive future. Learn more at: https://new.nsf.gov/funding/opportunities/design-environmental-sustainability-computing-desc
The U.S. National Science Foundation propels the nation forward by advancing fundamental research in all fields of science and engineering. NSF supports research and people by providing facilities, instruments and funding to support their ingenuity and sustain the U.S. as a global leader in research and innovation. With a fiscal year 2023 budget of $9.5 billion, NSF funds reach all 50 states through grants to nearly 2,000 colleges, universities and institutions. Each year, NSF receives more than 40,000 competitive proposals and makes about 11,000 new awards. Those awards include support for cooperative research with industry, Arctic and Antarctic research and operations, and U.S. participation in international scientific efforts.