An official website of the United States government
Environmentally-powered computer systems operate on renewable energy harvested from their environment, such as solar or wind, and stored in batteries. While harvesting environmental energy has long been necessary for small-scale embedded systems without access to external power sources, it is also increasingly important in designing sustainable larger-scale systems for edge applications. For sustained operations, such systems must consider not only the electrical energy but also the thermal energy available in the environment in their design and operation. Unfortunately, prior work generally ignores the impact of thermal effects, and instead implicitly assumes ideal temperatures. To address the problem, we develop a thermodynamic model that captures the interplay of elec- trical and thermal energy in environmentally-powered computer systems. The model captures the effect of environmental condi- tions, the system’s physical properties, and workload scheduling on performance. In evaluating our model, we distill the thermal effects that impact these systems using a small-scale prototype and a programmable incubator. We then leverage our model to show how considering these thermal effects in designing and operating environmentally-powered computer systems of varying scales can improve their energy-efficiency, performance, and availability.
more »
« less
This content will become publicly available on July 1, 2026
From Component to System: Rethinking Edge Computing Design through a Carbon-Aware Lens
As edge devices see increasing adoption across a wide range of applications, understanding their environmental impact has become increasingly urgent. Unlike cloud systems, edge deployments consist of tightly integrated microcontrollers, sensors, and energy sources that collectively shape their carbon footprint. In this paper, we present a carbon-aware design framework tailored to embedded edge systems. We analyze the embodied emissions of several off-the-shelf microcontroller boards and peripheral components and examine how deployment context—such as workload type, power source, and usage duration—alters the carbon-optimal configuration. Through empirical case studies comparing battery- and solar-powered scenarios, we find that the lowest-emission choice is often workload- and context-specific, challenging assumptions that energy-efficient or renewable powered systems are always the most sustainable. Our results highlight the need for fine-grained, system-level reasoning when designing for sustainability at the edge and provide actionable insights for researchers and practitioners seeking to reduce the carbon cost of future deployments.
more »
« less
- Award ID(s):
- 2324861
- PAR ID:
- 10648947
- Publisher / Repository:
- ACM
- Date Published:
- Journal Name:
- ACM SIGEnergy Energy Informatics Review
- Volume:
- 5
- Issue:
- 2
- ISSN:
- 2770-5331
- Page Range / eLocation ID:
- 125 to 131
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Extreme Edge Computing (EEC) promotes sustainable computing by reducing reliance on centralized data centres and decreasing their environmental impact. By using extreme edge devices to handle computing requests, the EEC reduces the energy demands for data transmission and execution, thereby reducing carbon footprints. However, EEC introduces challenges due to the mobile, heterogeneous, and resource-limited nature of these devices. Additionally, tasks are often complex and interdependent, complicating offloading and workload orchestration. The dynamicity of EEC systems, where both task generation and resources can be mobile, alongside task inter-dependencies, escalates the complexity of task offloading and workload management. To tackle these complexities, task partitioning emerges as a viable strategy. Moreover, in dynamic edge computing scenarios, resource demand remains unpredictable, emphasizing the critical need to optimize resource utilization efficiently. In this article, we investigate the problem of tasks with inter-dependencies offloading in an EEC environment where mobile and resource-constrained edge devices are employed as computing resources. In this regard, a partitioning-based Deep Reinforcement Learning (DRL) for Dependent sub-Task Orchestration (DeTOrch) model is proposed. DeTOrch uses a state-of-the-art partitioning method for decomposing tasks and proposes a novel mobility task-orchestration mechanism to minimize the task completion time and maximize the use of edge devices’ resource. The simulation results show that the proposed model can significantly improve the task success rate and decrease task completion time. In addition, in various scenarios with different levels of mobility, the proposed model outperforms the baselines while utilizing the resource of edge devices.more » « less
-
Generative AI, exemplified in ChatGPT, Dall-E 2, and Stable Diffusion, are exciting new applications consuming growing quantities of computing. We study the compute, energy, and carbon impacts of generative AI inference. Using ChatGPT as an exemplar, we create a workload model and compare request direction approaches (Local, Balance, CarbonMin), assessing their power use and carbon impacts. Our workload model shows that for ChatGPT-like services, in- ference dominates emissions, in one year producing 25x the carbon-emissions of training GPT-3. The workload model characterizes user experience, and experiments show that carbon emissions-aware algorithms (CarbonMin) can both maintain user experience and reduce carbon emissions dramatically (35%). We also consider a future scenario (2035 workload and power grids), and show that CarbonMin can reduce emissions by 56%. In both cases, the key is intelligent direction of requests to locations with low-carbon power. Combined with hardware technology advances, CarbonMin can keep emissions increase to only 20% compared to 2022 levels for 55x greater workload. Finally we consider datacenter headroom to increase effectiveness of shifting. With headroom, CarbonMin reduces 2035 emissions by 71%.more » « less
-
The low-latency requirements of connected electric vehicles and their increasing computing needs have led to the necessity to move computational nodes from the cloud data centers to edge nodes such as road-side units (RSU). However, offloading the workload of all the vehicles to RSUs may not scale well to an increasing number of vehicles and workloads. To solve this problem, computing nodes can be installed directly on the smart vehicles, so that each vehicle can execute the heavy workload locally, thus forming a vehicular edge computing system. On the other hand, these computational nodes may drain a considerable amount of energy in electric vehicles. It is therefore important to manage the resources of connected electric vehicles to minimize their energy consumption. In this paper, we propose an algorithm that manages the computing nodes of connected electric vehicles for minimized energy consumption. The algorithm achieves energy savings for connected electric vehicles by exploiting the discrete settings of computational power for various performance levels. We evaluate the proposed algorithm and show that it considerably reduces the vehicles' computational energy consumption compared to state-of-the-art baselines. Specifically, our algorithm achieves 15-85% energy savings compared to a baseline that executes workload locally and an average of 51% energy savings compared to a baseline that offloads vehicles' workloads only to RSUs.more » « less
-
Model-serving systems expose machine learning (ML) models to applications programmatically via a high-level API. Cloud plat- forms use these systems to mask the complexities of optimally managing resources and servicing inference requests across multi- ple applications. Model serving at the edge is now also becoming increasingly important to support inference workloads with tight latency requirements. However, edge model serving differs substan- tially from cloud model serving in its latency, energy, and accuracy constraints: these systems must support multiple applications with widely different latency and accuracy requirements on embedded edge accelerators with limited computational and energy resources. To address the problem, this paper presents Dělen,1 a flexible and adaptive model-serving system for multi-tenant edge AI. Dělen exposes a high-level API that enables individual edge applications to specify a bound at runtime on the latency, accuracy, or energy of their inference requests. We efficiently implement Dělen using conditional execution in multi-exit deep neural networks (DNNs), which enables granular control over inference requests, and evalu- ate it on a resource-constrained Jetson Nano edge accelerator. We evaluate Dělen flexibility by implementing state-of-the-art adapta- tion policies using Dělen’s API, and evaluate its adaptability under different workload dynamics and goals when running single and multiple applications.more » « less
