Quantifying Carbon Emission Footprints in Distributed Cloud Architectures via Runtime Telemetry Tracking

International Journal of Computer Science and Technology | Vol. 5, No. 3, March 2024 | pp. 1–8

DOI: 10.46882/2024/IJCST/000230

Article Type: Original Research Paper

Title: Quantifying Carbon Emission Footprints in Distributed Cloud Architectures via Runtime Telemetry Tracking

Names of Authors: Elena R. Rostova¹, Jean-Pierre Dubois²

Authors’ Affiliations: ¹Cloud Metrics Division, Global Tech Systems, Stockholm, Sweden; ²Department of Computer Science, Swiss Federal Institute of Technology, Lausanne, Switzerland

Abstract: Large-scale cloud data centers consume significant electrical energy, contributing to global carbon emissions. However, existing sustainability metrics estimate carbon impacts using static server configurations, failing to capture dynamic load variations or localized grid energy mixes. This paper presents an automated runtime monitoring framework named Eco-Track that quantifies carbon emission footprints at the individual container level. Eco-Track tracks server hardware activities, including CPU instruction counts, memory cycles, and thermal profiles. It cross-references this real-time consumption data with regional grid carbon intensity tracking feeds. The framework automatically moves low-priority processing tasks to data center zones operating on green energy sources. We tested the tracking tool across a multi-region distributed cloud infrastructure. The tracking results show that Eco-Track accurately measures individual software task carbon footprints with a margin of error under ±2.5%. Implementing our automated task migration loop reduced total operational carbon impacts by 18.6% without degrading application performance metrics.

Keywords: Green Computing, Cloud Data Centers, Carbon Footprint Tracking, Runtime Telemetry, Container Metrics, Sustainable Software

Manuscript Timeline: Received: August 25, 2023; Revised: October 30, 2023; Accepted: December 12, 2023; Published: March 06, 2024

Citation: Rostova, E. R., & Dubois, J. -P. (2024). Quantifying Carbon Emission Footprints in Distributed Cloud Architectures via Runtime Telemetry Tracking. International Journal of Computer Science and Technology, 5(3), 1–8. DOI: 10.46882/2024/IJCST/000230

Subscribe to read the full article: https://internationalscholarsjournals.org/subscribe-to-read