Datatracker: https://datatracker.ietf.org/rg/sustain/about/
ICS: https://datatracker.ietf.org/meeting/123/session/34419.ics
When: 17:00 - 19:00 CEST (15:00 - 17:00 UTC)
Room: Tapices -- https://datatracker.ietf.org/meeting/123/floor-plan
Co-chairs: Ali Rezaki (ali.rezaki@nokia.com), Eve Schooler (eve.schooler@gmail.com), Michael Welzl (michawe@ifi.uio.no)
Materials: https://datatracker.ietf.org/meeting/123/session/sustain
Note taking: https://notes.ietf.org/notes-ietf-123-sustain
Remote participation: https://meetings.conf.meetecho.com/ietf123/?session=34419
Local onsite tool: https://meetings.conf.meetecho.com/onsite123/?session=34419
Chat: https://zulip.ietf.org/#narrow/stream/sustain
Recording: (YouTube available post session)
REMINDER to Presenters: the duration times include Q&A
Note takers: Your names goes here!
Presenters: Ali Rezaki (Nokia), Eve Schooler (U Oxford), & Michael Welzl (U Oslo)
Charter: https://datatracker.ietf.org/doc/charter-irtf-sustain/
Presenter: Hermann de Meer (U Passau)
Contributors: Michael Lechl and Alexander Kilian (U Passau)
Abstract:
* Energy and ICT systems are increasingly interdependent, requiring integrated solutions for sustainable and stable operation. This talk presents approaches to unlock flexibility potentials from data centers and multi-use battery storage systems, considering energy efficiency from a local and grid perspective. For example, high-performance computing data centers can ship workloads based on renewable energy availability, reducing curtailment and enabling lowcarbon operation. It is depicted how wind-powered data centers and artificial intelligence can support intelligent, sustainable workload distribution. Multi-use battery storage systems complement this by providing grid support and balancing volatile renewable generation. Furthermore, co-locating data centers with battery storage systems at renewable generation sites allow for a dynamic level of Quality of Service (QoS) and Quality of Grid Service (QoGS) by introducing Green Service Level Agreements (GreenSLAs) and Green Supply Demand Agreements (GreenSDAs), for example. This integrated strategy supports both market and system needs while offering a pathway to dependable energy-ICT infrastructures. The presented use case shows that standards need to be established to realize energy efficiency and sustainability from a local perspective and a power grid perspective in decentralized systems consisting of consuming and generating entities.
Related work:
* A. Berl, E. Gelenbe, M. Di Girolamo, G. Giuliani, H. de Meer, M. Quan Dang and K. Pentikousis, "Energy-Efficient Cloud Computing", The Computer journal, vol. 53, no. 7, pp. 1045-1051, Aug. 2009. Oxford University Press. DOI: 10.1093/comjnl/bxp080
* R. Basmadjian, J. F. Botero, G. Giuliani, X. Hesselbach, S. Klingert and H. de Meer, "Making Data Centers Fit for Demand Response: Introducing GreenSDA and GreenSLA Contracts", IEEE Transactions on Smart Grid, vol. 9, no. 4, pp. 3453-3464, Nov. 2018. DOI: 10.1109/TSG.2016.2632526
* A. Kilian, H. de Meer and G. Schomaker, "Energy-Optimized Supercomputer Networks Using Wind Energy", Communications of the ACM (CACM), vol. 68, no. 7, pp. 74-79, Jul. 2025. ACM. DOI: 10.1145/3725981
* M. Lechl, T. Fürmann, H. de Meer and A. Weidlich, "A review of models for energy system flexibility requirements and potentials using the new FLEXBLOX taxonomy", Renewable and Sustainable Energy Reviews, vol. 184, pp. 1-19, Sep. 2023. Elsevier. DOI: 10.1016/j.rser.2023.113570
* M. Lechl, A. Kilian and H. de Meer, "Uncertainty-Aware Scheduling of Multi-Use Battery Storage Systems" in 16th ACM Int’l. Conf. on Future and Sustainable Energy Systems (e-Energy '25), June 17-20, 2025, Rotterdam, Netherlands, ACM, Jun. 2025. pp. 243-256. DOI: 10.1145/3679240.3734607
* M. Lechl, H. de Meer and T. Fürmann, "A Stochastic Flexibility Calculus for Uncertainty-Aware Energy Flexibility Management", Applied Energy, vol. 379, pp. 1 16, Feb. 2025. Elsevier. DOI: 10.1016/j.apenergy.2024.124907
Presenter: Noa Zilberman (U Oxford)
Abstract:
* Leveraging the temporal and spatial variability of carbon intensity presents a significant opportunity to reduce the Internet's carbon emissions. However, maximizing the impact of carbon awareness requires addressing specific challenges by the IRTF community. In this talk, I will explore how carbon awareness can drive the reduction of operational carbon emissions across the network. I'll discuss key pain points that can be helped by technological advancements, but also necessitate community consensus and standardization. As processing operations increasingly shift from centralized cloud environments to the distributed edge, I will also introduce emerging concerns and discuss plans for deploying carbon-aware network solutions.
Related work:
* Noa Zilberman, Eve M. Schooler, Uri Cummings, Rajit Manohar, Dawn Nafus, Robert Soulé, and Rick Taylor. Toward Carbon-Aware Networking. HotCarbon'22, July 2022
* Sawsan El Zahr, Paul Gunning and Noa Zilberman. Exploring the Benefits of Carbon-Aware Routing. ACM CoNEXT and Proceedings of the ACM on Networking (PACMNET), December 2023
* Sawsan El-Zahr, William Nathan, Noa Zilberman. Carbon-intelligent content scheduling in CDNs, ACM/IRTF ANRW, July 2025
* Alexander Clemm, Dirk Kutscher, Michael Welzl, Cedric Westphal, Noa Zilberman, and Simone Ferlin-Reiter. Greening Networking: Toward a Net Zero Internet (Dagstuhl Seminar 24402)". Dagstuhl Reports, Volume 14, Issue 9, 2025
* Yuta Tokusashi, Huynh Tu Dang, Fernando Pedone, Robert Soulé and Noa Zilberman, The Case For In Network Computing On Demand, Eurosys 2019
* National Engineering Policy Centre (NEPC), Engineering Responsible AI: foundations for environmentally sustainable AI, Feb 2025
* SmartEdge Project, Innovative Smart Edge Solutions, 2025
Presenter: Hunter Vaughan (Emerson College)
Paper: Report on Best Practices in Subsea Telecommunications Sustainability
Publications Archive: Sustainable Subsea Networks
Abstract:
* In this talk, I will discuss the challenges to environmental and social sustainability presented by digital infrastructure growth in an era of accelerated climate crisis. This will include a summary of our two-stage four-year grant work on the Sustainable Subsea Networks project; my own specific reflections on attempting to lead boundary-spanning collaborative research with industry, local communities, and experts from marine ecology to renewable energy; and expansive consideration on the future costs and consequences of infrastructural growth to cater to the heavy data needs of smart cities and high computational machine learning (eg. "AI", large language models such as Chat GPT, etc).
Presenter: Jan Lindblad (All for Eco AB)
Abstract:
* This presentation proposes four areas of internet research related to energy management and GHG reporting. Due to the lack of standards, organizations invent their own methods for assessing their networking footprint. Often, this does not end well. Attribution of scope 3 emissions is done ad-hoc. While addressing these questions, the reporting stacks of governments around the world need to be considered. Finally, a device-type neutral and vendor agnostic language for expressing power saving is needed.
Presenter: Jesus Alberto Omaña Iglesias (Telefonica)
Contributors: Andra Lutu (Telefonica)
Abstract:
* Reducing energy consumption is a primary goal for the mobile telecommunication industry, with strong environmental and economic implications. The main target for savings is the Radio Access Network (RAN), which is responsible for more than 70% of the total energy costs incurred by operators. Lowering energy costs at the RAN is possible by reducing the number of active carriers at off-peak locations and times where the demand can be served with a lower capacity than deployed. While the scientific community has been proposing a plethora of complex solutions to switch off underutilized carriers, production networks largely rely nowadays on threshold-based strategies that run at individual RAN equipment and are typically enabled only overnight. Moreover, there are no real-world evaluations of the effectiveness of carrier switch-off approaches in reducing energy consumption or their impact on the end users. In this talk, we present the results of benchmarking five fixed-threshold-based cell sleep policies deployed in a production network serving large geographical regions. The study provides unprecedented insights on industry-grade RAN sustainability at scale, in terms of actual energy savings and trade-offs with user experience. Our insights suggest that the capability of the tested policies in reducing the energy costs hits a clear barrier if no degradation is admissible for any user, and provides a strong empirical basis in support of more flexible approaches to save energy at the RAN.
Related Work:
* O. E. Martínez-Durive, J. Suárez-Varela, J. O. Iglesias, A. Lutu, and M. Fiore, An Evaluation of RAN Sustainability Strategies in Production Networks, IEEE INFOCOM 2025
* M. Kalntis, A. Lutu, J. O. Iglesias, F. A. Kuipers, and G. Iosifidis, Smooth Handovers via Smoothed Online Learning, IEEE INFOCOM 2025
* M. Kalntis, J. Suárez-Varela, J. O. Iglesias, A. K. Bhattacharjee, G. Iosifidis, F. A. Kuipers, and A. Lutu, Through the Telco Lens: A Countrywide Empirical Study of Cellular Handovers, ACM IMC ’24
Presenter: Wen Cai (U Oslo)
Abstract
* Energy-efficient computing requires access to carbon intensity information, which forms the basis for carbon-aware routing, load shifting, network optimization, and the selection of locations for processing and storage. Existing APIs and databases — such as ElectricityMap — divide the world into large areas like “Bidding Zones” (European regions defined by ENTSO-E), offering only coarse granularity in regional carbon intensity.
However, corporations and organizations need more than a regional carbon-intensity map. We therefore propose a freely available carbon map with finer granularity: a much more detailed, dynamic database at the city level for each country.
Our methodology estimates carbon intensity by taking into account city-level population, local energy consumption, and energy production, using publicly available data to generate a high-resolution global carbon intensity map.
Presenter: Alex Clemm
Paper: Towards Sustainable Networking: Unveiling Energy Efficiency Through Hop and Path Efficiency Indicators in Computer Networks
Internet-Draft: draft-cxx-ippm-ioamaggr-03
Abstract:
* Observability of metrics relating to energy use is a key enabler for applications that aim to improve the sustainability of networks being operated. While the main focus is typically on metrics provided by individual devices and their components, some metrics can also be applied to paths in a network to allow, for example, for smarter path selection decisions. However, the collection of such data is not trivial today as it requires correlation of multiple data points collected individually across devices.
This presentation motivates the introduction and provides an overview of Green Path Metrics related to the energy use and efficiency of network traffic traversing multiple devices. It provides an overview of a recently proposed Aggregation Trace Option, an extension of iOAM technology that allows to aggregate path metrics in a network without needing to contend with issues such as the need to correlate and postprocess raw data or packet sizes that grow with path lengths. A PoC that uses this new protocol to obtain sustainability path metrics is briefly presented.
Presenter: Jari Arkko (Ericsson)
Internet-Draft: draft-various-eimpact-arch-considerations-01
Abstract:
* The question of how architecture can influence sustainability has been the topic for a design team in the IETF from the beginning of the year. The early results were reported at length in the SUSTAIN RG meeting in March. This talk reports briefly the most recent updates, and calls for review and action. In particular, we want to highlight the call for research on some aspects of this, such as understanding the interaction between transport protocol behavior and network node sleep modes.