Program

Workshop opening and awards

Best Paper Award
A Sleep Study for ISP Networks: Evaluating Link Sleeping on Real World Data

Keynote: Quantifying the carbon footprint of AI and computing: Past, Present, and Future

Udit Gupta (Cornell)

Computing has a substantial environmental impact, with far-reaching consequences. Information and Communication Technologies (ICT) alone contribute to 4% of global carbon emissions, equivalent to the aviation industry’s carbon footprint. This impact is poised to increase as the demand for computing continues to surge, particularly at the edge and within massive hyperscale data centers. Creating environmentally sustainable computing solutions presents unique challenges that go beyond energy efficiency optimization. It necessitates a holistic approach that considers various factors. For example, sustainable computing devices must account for the carbon emissions generated during the manufacturing process called embodied emissions, the influence of renewable energy sources that fluctuate by location and time on a system’s operational emissions, and the intricate interplay between embodied and operational emissions. However, to enable sustainability-driven system design and optimization, we must first be able to quantify the carbon footprint of computing systems. In this presentation, we will delve into recent breakthroughs and the research challenges we confront as we develop tools and methodologies to quantify the carbon footprint of computing systems across their lifetime, from hardware manufacturing to operational use. We will introduce new carbon accounting models, highlight their application to pertinent domains like AI to promote developing sustainable AI solutions, as well as the road ahead in enabling sustainability as a first order design tenet alongside performance, efficiency, and cost.

Session 1: Carbon-Aware Computing and Data Centers

Chair: Simon Peter (UW)

Understanding the Operational Carbon Footprint of Storage Reliability and Management

  • Varsha Rao (University of Chicago)
  • Andrew A. Chien (University of Chicago)

Data-driven Algorithm Selection for Carbon-Aware Scheduling

  • Roozbeh Bostandoost (University of Massachusetts Amherst)
  • Walid A. Hanafy (University of Massachusetts Amherst)
  • Adam Lechowicz (University of Massachusetts Amherst)
  • Noman Bashir (Massachusetts Institute of Technology)
  • Prashant Shenoy (University of Massachusetts Amherst)
  • Mohammad Hajiesmaili (University of Massachusetts Amherst)

A Call for Research on Storage Emissions

  • Sara McAllister (Carnegie Mellon University)
  • Fiodar Kazhamiaka (Microsoft Azure)
  • Daniel S. Berger (Microsoft Azure)
  • Rodrigo Fonseca (Microsoft Azure)
  • Kali Frost (Microsoft Azure)
  • Aaron Ogus (Microsoft Azure)
  • Maneesh Sah (Microsoft Azure)
  • Ricardo Bianchini (Microsoft Azure)
  • George Amvrosiadis (Carnegie Mellon University)
  • Nathan Beckmann (Carnegie Mellon University)
  • Gregory R. Ganger (Carnegie Mellon University)

Uncertainty-Aware Decarbonization for Datacenters

  • Amy Li (University of Waterloo)
  • Sihang Liu (University of Waterloo)
  • Yi Ding (Purdue University)

Geographical Server Relocation: Opportunities and Challenges

  • Yejia Liu (UC Riverside)
  • Pengfei Li (UC Riverside)
  • Daniel Wong (University of California, Riverside)
  • Shaolei Ren (UC Riverside)

Session 2: Sustainable AI and Streaming

Chair: Esha Choukse (Microsoft)

Towards Sustainable Large Language Model Serving

  • Sophia Nguyen (University of Waterloo)
  • Beihao Zhou (University of Waterloo)
  • Yi Ding (Purdue University)
  • Sihang Liu (University of Waterloo)

Carbon in Motion: Characterizing Open-Sora on the Sustainability of Generative AI for Video Generation

  • Baolin Li (Northeastern University)
  • Yankai Jiang (Northeastern University)
  • Devesh Tiwari (Northeastern University)

Towards Carbon-efficient LLM Life Cycle

  • Yueying Li (Cornell University)
  • Udit Gupta (Cornell University)
  • Omer Graif (Cornell University)

Streaming the future of sustainability

  • Ryan Hardesty Lewis (Cornell Tech)
  • Alex Conway (Cornell Tech)

Session 3: Energy and Power

Chair: Timo Hoenig (RUB)

Energy-Aware Process Scheduling in Linux

  • Feitong Qiao (Columbia University)
  • Yiming Fang (Columbia University)
  • Asaf Cidon (Columbia University)

A Sleep Study for ISP Networks: Evaluating Link Sleeping on Real World Data

Best Paper Award

  • Lukas Roellin (ETH Zürich)
  • Romain Jacob (ETH Zürich)
  • Laurent Vanbever (ETH Zürich)

Learning a Data Center Model for Efficient Demand Response

  • Quentin Clark (Boston University)
  • Fatih Acun (Boston University)
  • Ioannis Ch. Paschalidis (Boston University)
  • Ayse K. Coskun (Boston University)

EMPower: The Case for a Cloud Power Control Plane

  • Jonggyu Park (University of Washington)
  • Theano Stavrinos (University of Washington)
  • Simon Peter (University of Washington)
  • Thomas Anderson (University of Washington)

Proactive Energy Management in Database Systems

  • Yuto Hayamizu (The University of Tokyo)
  • Masaru Kitsuregawa (The University of Tokyo)
  • Kazuo Goda (The University of Tokyo)

Session 4: Sustainable Hardware and Cloud Services

Chair: George Porter (UCSD)

Lovelock: Towards Smart NIC-hosted Clusters

  • Seo Jin Park (University of Southern California)
  • Ramesh Govindan (University of Southern California)
  • Kai Shen (Google)
  • David Culler (Google)
  • Fatma Özcan (Google)
  • Geon-Woo Kim (The University of Texas at Austin)
  • Hank Levy (Google and University of Washington)

Bridging the Sustainability Gap in Serverless through Observability and Carbon-Aware Pricing

  • Changyuan Lin (University of British Columbia)
  • Mohammad Shahrad (University of British Columbia)

Understanding the Implications of Uncertainty in Embodied Carbon Models for Sustainable Computing

  • Anvita Bhagavathula (Cornell University)
  • Leo Han (Cornell University)
  • Udit Gupta (Cornell University)

Towards Game-Theoretic Approaches to Attributing Carbon in Cloud Data Centers

  • Leo Han (Cornell)
  • Jash Kakadia (University of Pennsylvania)
  • Benjamin C Lee (University of Pennsylvania)
  • Udit Gupta (Cornell)

Session 5: Open Discussion

Chair: Simon Peter (UW) and Esha Choukse (Microsoft)

Online Program

CO2CoDe: Towards Carbon-Aware Hardware/Software Co-Design for Intermittently-Powered Embedded Systems

  • Phillip Raffeck (Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU))
  • Sven Posner (Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU))
  • Peter Wägemann (Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU))

Vessim: A Testbed for Carbon-Aware Applications and Systems

  • Philipp Wiesner (Technische Universität Berlin)
  • Ilja Behnke (Technische Universität Berlin)
  • Paul Kilian (Technische Universität Berlin)
  • Marvin Steinke (Technische Universität Berlin)
  • Odej Kao (Technische Universität Berlin)

Offline Energy-Optimal LLM Serving: Workload-Based Energy Models for LLM Inference on Heterogeneous Systems

  • Grant Wilkins (University of Cambridge)
  • Srinivasan Keshav (University of Cambridge)
  • Richard Mortier (University of Cambridge)

Improving Carbon Emissions of Federated Large Language Model Inference through Classification of Task-Specificity

  • Geerd-Dietger Hoffmann (Green Coding Solutions GmbH)
  • Verena Majuntke (HTW Berlin)

BoaviztAPI: A Bottom-Up Model to Assess the Environmental Impacts of Cloud Services

  • Thibault Simon (Orange Labs, Univ. Lille, Inria, CNRS, UMR 9189 CRIStAL)
  • David Ekchajzer (Hubblo, Université Paris-Saclay, Univ Evry, IMT-BS, LITEM,)
  • Adrien Berthelot (ENS de Lyon, OCTO Technology, Inria)
  • Eric Fourboul (Hubblo, Boavizta)
  • Samuel Rince (Alygne, Boavizta)
  • Romain Rouvoy (Univ. Lille, Inria, CNRS, UMR 9189 CRIStAL)

Understanding and Mitigating Webpage Data Bloat: Causes and Preventive Measures

  • Aleksi Saarinen (Aalto University)
  • Muhammad Zubair Farooqi (Aalto University)
  • Matti Pärssinen (Aalto University)
  • Jukka Manner (Aalto University)