Using Budgets to Reduce Application Emissions
Leo Wilhelm Lierse, Mahyar Tourchi Moghaddam, Sebastian Werner

TL;DR
This paper introduces emissions budgets as a flexible method for managing application emissions, allowing dynamic adaptation to grid carbon intensity and improving task fulfillment in variable energy conditions.
Contribution
It proposes a novel emissions budget approach that replaces fixed emission rates, enabling applications to adapt dynamically and maintain emission limits while optimizing performance.
Findings
Budget-based management improves task fulfillment by up to 36% in variable grids.
Budgets match fixed rate performance in stable grids, ensuring safety.
Simulation with real-world data demonstrates practical effectiveness.
Abstract
As carbon pricing mechanisms like the EU Emissions Trading System are set to increase prices of energy consumption, software architects face growing pressure to design applications that operate within financially predictable emission constraints. Existing approaches typically enforce rigid per-interval emission rates, which prove unsuitable in electrical grids with highly dynamic carbon intensity, which is common in grids with growing renewable energy adoption. We propose the use of emissions budgets, an approach that replaces fixed emission rates with time-bound budgets, enabling applications to dynamically save unused emission allowances during low carbon intensity periods and expend them during high carbon intensity periods. We describe emissions-aware applications using a MAPE-K feedback loop that continuously monitors application power consumption and grid carbon intensity, then…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
