TCSE logo 
 Sigsoft logo
Sustainability badge

While Generative AI stands to be one of the fastest adopted technologies ever, studies have made evident that the usage of Large Language Models (LLMs) puts significant burden on energy grids and our environment. It may prove a hindrance to the Sustainability goals of any organization. A crucial step in any Sustainability strategy is monitoring or estimating the energy consumption of various components. While there exist multiple tools for monitoring energy consumption, there is a dearth of tools/frameworks for estimating the consumption or carbon emissions. Current drawbacks of both monitoring and estimation tools include high input data points, intrusive nature, high error margin, etc. We posit that leveraging emerging LLM benchmarks and related data points can help overcome aforementioned challenges while balancing accuracy of the emission estimations. To that extent, we discuss the challenges of current approaches and present our evolving framework, R-ICE which estimates prompt level inference carbon emissions by leveraging existing state-of-the-art(SOTA) benchmark. This direction provides a more practical and non-intrusive way to enable emerging use-cases like dynamic LLM routing, carbon accounting, etc. Our promising validation results suggest that benchmark-based modelling holds great potential for inference emission estimation and warrants further exploration from the scientific community.

Tue 29 Apr

Displayed time zone: Eastern Time (US & Canada) change

11:00 - 12:30
Session 2: Pitch Session 2 (7-minute pitch of each paper and 3-minute question/comment)GREENS at 203
Chair(s): Elisa Yumi Nakagawa University of São Paulo
11:00
10m
Talk
Generating Energy-efficient code with LLMs
GREENS
Tom Cappendijk University of Amsterdam, Pepijn de Reus University of Amsterdam, Ana Oprescu University of Amsterdam
11:10
10m
Talk
Prompt engineering and its implications on the energy consumption of Large Language Models
GREENS
Riccardo Rubei University of L'Aquila, Aicha Moussaid University of L'Aquila (Italy), Claudio Di Sipio University of l'Aquila, Davide Di Ruscio University of L'Aquila
11:20
10m
Talk
Breaking the ICE: Exploring promises and challenges of benchmarks for Inference Carbon & Energy estimation for LLMs
GREENS
Samarth Sikand Accenture Labs, Rohit Mehra Accenture Labs, Priyavanshi Pathania Accenture Labs, Nikhil Bamby Accenture Labs, Vibhu Saujanya Sharma Accenture Labs, Vikrant Kaulgud Accenture Labs, India, Sanjay Podder Accenture, Adam P. Burden Accenture
11:30
10m
Talk
Calculating Software’s Energy Use and Carbon Emissions: A Survey of the State of Art, Challenges, and the Way Ahead
GREENS
Priyavanshi Pathania Accenture Labs, Nikhil Bamby Accenture Labs, Rohit Mehra Accenture Labs, Samarth Sikand Accenture Labs, Vibhu Saujanya Sharma Accenture Labs, Vikrant Kaulgud Accenture Labs, India, Sanjay Podder Accenture, Adam P. Burden Accenture
11:40
10m
Talk
Responsible and Sustainable AI: Considering Energy Consumption in Automated Text Classification Evaluation Tasks
GREENS
Angelika Kaplan Karlsruhe Institute of Technology (KIT), Jan Keim Karlsruhe Institute of Technology (KIT), Lukas Greiner Karlsruhe Institute of Technology (KIT), Ralf Sieger FZI Research Center for Information Technology, Raffaela Mirandola Karlsruhe Institute of Technology (KIT), Ralf Reussner Karlsruhe Institute of Technology (KIT) and FZI - Research Center for Information Technology (FZI)
DOI Pre-print
11:50
10m
Talk
Mapping of the system of software-related emissions and shared responsibilities
GREENS
Laura Partanen LUT University, Antti Sipilä LUT University, Sanaul Haque LUT University, Jari Porras LUT University
12:00
10m
Talk
PowerLetrics: An Open-Source Framework for Power and Energy Metrics for Linux
GREENS
Geerd-Dietger Hoffmann Employed by Green Coding Solutions, Verena Majuntke HTW Berlin
12:10
10m
Talk
Echoes of the Future: Designing a Game for Green Software Engineering
GREENS
Georgia Samaritaki University of Amsterdam, Humeyra Tugce Yavuz University of Amsterdam, Daphnee Chabal University of Amsterdam, Ana Oprescu University of Amsterdam
12:20
10m
Other
Final question/comment filling and posting (via Miro)
GREENS

:
:
:
: