HotCarbon’2023

Session 1: Systems

This session focuses on systems energy and carbon quantification and optimization, including CPU, GPU, DRAMs, their configuration, tools, and accuracy evaluation.

The Case of Unsustainable CPU Affinity

Jiechen Zhao (University of Toronto); Katie Lim, Thomas Anderson (University of Washington); Natalie Enright Jerger (University of Toronto)

carbond: An Operating-System Daemon for Carbon Awareness

Andreas Schmidt, Gregory Stock, Robin Ohs (Saarland Informatics Campus); Luis Gerhorst (Friedrich-Alexander-Universität Erlangen-Nürnberg); Benedict Herzog, Timo Hönig (Ruhr-Universität Bochum (RUB))

Carbon-Aware Memory Placement

Sven Köhler (Hasso-Plattner-Institut (HPI)); Benedict Herzog, Henriette Hofmeier, Manuel Vögele (Ruhr-Universität Bochum (RUB)); Lukas Wenzel, Andreas Polze (Hasso-Plattner-Institut (HPI)); Timo Hönig (Ruhr-Universität Bochum (RUB))

EnergAt: Fine-Grained Energy Attribution for Multi-Tenancy

Hongyu Hè, Michal Friedman (ETH); Theodoros Rekatsinas (Apple Inc.)

Towards Application Centric Carbon Emission Management

Sudarsun Kannan, Ulrich Kremer (Rutgers University)

Track 1 combined Q&A (video)

Session 2: Multi Clouds: Unique challenges and opportunities

This session focuses on the unique challenges and potential research opportunities related to cloud environments.

Bringing Carbon Awareness to Multi-cloud Application Delivery

Diptyaroop Maji (University of Massachusetts Amherst); Ben Pfaff (Feldera); Rajagopal Sreenivasan (VMware); Colleen Josephson (VMware and UC Santa Cruz); Sreeram Iyer, Zhelong Pan, Vipin Padmam Ramesh, Victor Firoiu (VMware); Ramesh K Sitaraman (UMass Amherst & Akamai Tech)

Myths and Experiences in Reducing Carbon for Cloud Platforms

Jialun Lyu (Microsoft and University of Toronto); Jaylen Wang (Microsoft and Carnegie Mellon University); Kali Frost (Microsoft Research); Chaojie Zhang, Celine Irvene, Esha Choukse, Rodrigo Fonseca, Ricardo Bianchini, Fiodar Kazhamiaka (Microsoft Azure); Daniel S. Berger (Microsoft Azure and University of Washington)

Peeling Back the Carbon Curtain: Carbon Optimization Challenges in Cloud Computing

Jaylen Wang (Carnegie Mellon University); Udit Gupta (Cornell University); Akshitha Sriraman (Carnegie Mellon University / Google)

Towards Carbon Footprint Management in Hybrid Multicloud

Rohan Arora (IBM Research); UmaMaheswari Devi (IBM Research - India); Tamar Eilam (IBM); Aanchal Goyal (IBM Research - India); Chandra Narayanaswami, Pritish Parida (IBM Research)

An Agile Pathway Towards Carbon-aware Clouds

Pratyush Patel, Theo Gregersen, Thomas Anderson (University of Washington)

Track 2 combined Q&A (video)

Session 3: Data and AI

This session focuses on the unique challenges and opportunities in addressing the rising energy and carbon cost of Data and AI.

Reducing the Carbon Impact of Generative AI Inference (today and in 2035)

Andrew A. Chien (University of Chicago and Argonne National Laboratory); Liuzixuan Lin, Hai Nguyen, Varsha Rao, Tristan Sharma, Rajini Wijayawardana (University of Chicago)

Carbon-Efficient Neural Architecture Search

Yiyang Zhao, Tian Guo (Worcester Polytechnic Institute)

Towards a Methodology and Framework for AI Sustainability Metrics

Tamar Eilam (IBM T. J. Watson Research Center); Pedro Bello-Maldonado (IBM Thomas J. Watson Research Center); Bishwaranjan Bhattacharjee (IBM T. J. Watson Research Center); Carlos Costa (IBM Thomas J. Watson Research Center); Eun Kyung Lee (IBM T. J. Watson Research Center); Asser Tantawi (IBM Thomas J. Watson Research Center)

Toward a Life Cycle Assessment for the Carbon Footprint of Data

Gabriel Mersy, Sanjay Krishnan (The University of Chicago)

Track 3 combined Q&A (video)

Session 4: The Life Cycle of Applications

This session focuses on tradeoffs across the life cycle such as embodied carbon calculation, tradeoffs between energy and carbon, and also, global policy issues.

On the Promise and Pitfalls of Optimizing Embodied Carbon

Noman Bashir, David Irwin, Prashant Shenoy (University of Massachusetts Amherst)

Carbon-Efficient Design Optimization for Computing Systems

Mariam Elgamal (Harvard University); Doug Carmean, Elnaz Ansari, Okay Zed, Ramesh Peri, Srilatha Manne, Udit Gupta (Meta); Gu-Yeon Wei (Harvard University); David Brooks (Harvard University/Meta); Gage Hills (Harvard University); Carole-Jean Wu (Meta)

Does rate adaptation at daily timescales make sense?

Romain Jacob, Jackie Lim, Laurent Vanbever (ETH Zürich)

Transient Internet of Things: Redesigning the Lifetime of Electronics for a More Sustainable Networked Environment

Tingyu Cheng (Georgia Institute of Technology); Gregory D. Abowd (Northeastern University); Hyunjoo Oh, Josiah Hester (Georgia Institute of Technology)

The War of the Efficiencies: Understanding the Tension between Carbon and Energy Optimization

Walid A. Hanafy, Roozbeh Bostandoost, Noman Bashir, David Irwin, Mohammad Hajiesmaili, Prashant Shenoy (University of Massachusetts Amherst)

When Does Saving Power Save the Planet?

Jackson Woodruff, David Schall, Michael O’Boyle (University of Edinburgh); Christopher Woodruff (University of Oxford)

Track 4 combined Q&A (video)

Panel: “Can AI be sustainable?”

Panelists: Martha Kim (Columbia Univ.), Ramya Raghavendra (Meta), Huamin Chen (RedHat), Andrew Chien (Univ. of Chicago), Sanjay Krishnan (Univ. of Chicago)

Link to panel slides

Link to the panel discussion (video)