Data centers consume roughly 1--2% of the world's electricity, with the majority of it attributed to compute, making the computing industry a substantial source of greenhouse gas emissions. Resources in data centers typically focus on providing high performance and availability, but the question of sustainability in managing these distributed resources often goes unnoticed over these other metrics. This problem will only exacerbate as the data center computing demand continues to increase. In this paper, we address the sustainability aspect of load balancing in VMware's Avi Global Server Load Balancer (GSLB). Our GSLB deployment spans data centers across geographies and clouds and relies on geographical proximity to shift client application requests to the closest data center. In this work, we enhance the GSLB service to additionally consider the real-time carbon intensity at each data center as a factor in making a load-balancing choice. Our carbon-aware prototype shows an average of 21% and a maximum of 51% reduction in carbon emissions while operating with an acceptable latency.

Human-caused climate degradation and the explosion of electronic waste have pushed the computing community to explore fundamental alternatives to the current battery-powered, over-provisioned ubiquitous computing devices that need constant replacement and recharging. Soil Microbial Fuel Cells (SMFCs) offer promise as a renewable energy source that is biocompatible and viable in difficult environments where traditional batteries and solar panels fall short. However, SMFC development is in its infancy, and challenges like robustness to environmental factors and low power output stymie efforts to implement real-world applications in terrestrial environments. This work details a 2-year iterative process that uncovers barriers to practical SMFC design for powering electronics, which we address through a mechanistic understanding of SMFC theory from the literature. We present nine months of deployment data gathered from four SMFC experiments exploring cell geometries, resulting in an improved SMFC that generates power across a wider soil moisture range. From these experiments, we extracted key lessons and a testing framework, assessed SMFC's field performance, contextualized improvements with emerging and existing computing systems, and demonstrated the improved SMFC powering a wireless sensor for soil moisture and touch sensing. We contribute our data, methodology, and designs to establish the foundation for a sustainable, soil-powered future.

Recent work shows the exciting potential for soil microbes as a renewable energy harvesting source. However, the choice of materials in microbial fuel cells (MFCs) significantly impacts where the energy comes from. MFCs with metallic anodes draw energy from both renewable bacterial activity and non-renewable galvanic corrosion of cell components. Previous studies do not analyze these two power sources separately. This letter clarifies the behavior of metallic MFCs by characterizing galvanic activity separately from biological activity. We find that the majority of energy attained from prior designs is most likely galvanic, not bacterial, and as a consequence is non-renewable. This letter further summarizes best practices for identifying and characterizing bacteria so as to better understand the performance of future MFCs.

Forest mapping provides critical observational data needed to understand the dynamics of forest environments. Notably, tree diameter at breast height (DBH) is a metric used to estimate forest biomass and carbon dioxide (${\mathrm{CO}}_2$) sequestration. Manual methods of forest mapping are labor intensive and time consuming, a bottleneck for large-scale mapping efforts. Automated mapping relies on acquiring dense forest reconstructions, typically in the form of point clouds. Terrestrial laser scanning (TLS) and mobile laser scanning (MLS) generate point clouds using expensive LiDAR sensing and have been used successfully to estimate tree diameter. Neural radiance fields (NeRFs) are an emergent technology enabling photorealistic, vision-based reconstruction by training a neural network on a sparse set of input views. In this paper, we present a comparison of MLS and NeRF forest reconstructions for the purpose of trunk diameter estimation in a mixed-evergreen Redwood forest. In addition, we propose an improved DBH-estimation method using convex-hull modeling. Using this approach, we achieved 1.68 cm RMSE (2.81%), which consistently outperformed standard cylinder modeling approaches.