Carbon Aware Computing
By 2030, we envision a world where intelligent edge devices will be ubiquitous, impacting our daily lives in various ways. Spanning a wide spectrum, from energy-harvesting devices (e.g., sensors, backscatter tags) to more sophisticated platforms including mobile phones, wearables, and even small drones, edge devices will provide accessibility and critical health and wellness applications to many communities. The increasing sophistication and capabilities of these devices, primarily fueled by advancements in artificial intelligence and hardware efficiency, will lead to a dramatic rise in the number of intelligent edge devices deployed.
Given their growing ubiquity, we must consider the environmental impact of edge devices. Information and communications technology is responsible for up to 4% of worldwide emissions, equivalent to the aviation industry’s carbon footprint. User and edge devices are responsible for more than one-third of this carbon footprint. This environmental impact is expected to worsen as the number of edge devices (i.e., IoT devices, sensors, consumer devices) is projected to increase from nearly 100 billion to trillions.
The significant improvements in performance and energy efficiency over the last two decades remain insufficient to reduce the environmental impact of computing devices. First, in terms of carbon emissions, the operational emissions resulting from energy consumption over a device’s lifetime are only a fraction of its impact; the embodied carbon emitted during the production of computing devices plays a more significant role. Moreover, most edge devices are equipped with computational resources and batteries that exceed their lifetime and use case needs, leading to significant amounts of wasted carbon emissions, particularly embodied emissions. Finally, the environmental impact of edge devices extends beyond carbon emissions to include e-waste from discarded electronics, the toxicity of materials used for integrated circuit (IC) and battery fabrication, and the water consumed during IC manufacturing.
This project seeks to make carbon and sustainability a first-order design parameter for future edge computing devices that range from tiny, energy-harvesting Internet of Things (IoT) devices to higher-performance consumer electronics. Delphi is a suite of carbon-aware design tools that consider factors like energy, e-waste, and water usage from the manufacturing of computational devices, as well as operation carbon factors of machine learning and software lifecycles. The framework covers sensors, computing, communication, and power, accommodating various edge devices, and includes probabilistic analysis of machine learning workflows design and their operational and embodied carbon impact.
The project’s research has three major tasks. First, quantifying device environmental impact by collecting a first-of-its-kind dataset via a state-of-the-art academic clean room, the Cornell Nanoscale Facility (CNF), with architectural carbon models for salient device components (e.g., processor, memory and storage, energy harvesting modules). This task integrates data into new foundational carbon models, guiding all research tasks. Second, tools for the design of systems with sustainability as a first-order design target, alongside performance and quality of service. The task develops Electronic Sustainability Records for devices on the Pareto-frontier to maintain system-specific sustainability ledgers to track environmental telemetry across the operational lifetime of devices. Finally, the third task develops runtime sustainability managers, including humans in the loop, to reduce device obsolescence. The software will gracefully degrade and upgrade system performance based on user choices, static lifetime requirements, and environmental factors. The comprehensive framework’s effectiveness is demonstrated through short-lived “Ephemeral devices” and lifelong companion health and wellness wearable devices, nicknamed the “Infinite Bit.” These two device archetypes provide a mechanism for continuous validation as the project matures.