pith. sign in

arxiv: 2605.24569 · v1 · pith:ET6XDR4Ynew · submitted 2026-05-23 · 💻 cs.DC · cs.PF

Energy-Aware Computing in the Year 2026

Roblex Nana Tchakoute , Claude Tadonki This is my paper

Pith reviewed 2026-06-30 12:17 UTC · model grok-4.3

classification 💻 cs.DC cs.PF
keywords energy-efficient computingHPCtaxonomysustainable computingvoltage scalingworkload consolidationfederated learningcooling
0
0 comments X

The pith

A holistic taxonomy organizes energy-efficient techniques across hardware, software, scheduling, and cooling for the cloud-edge-HPC continuum.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper surveys recent work on reducing power use in high-performance computing and large-scale AI systems, where energy has become a central limit on scaling. It groups these efforts into a single taxonomy that spans hardware choices, measurement tools, code optimizations, task scheduling, voltage adjustments, workload packing, federated learning, and cooling methods. The taxonomy is meant to reveal gaps and guide future work toward lower carbon footprints and sustainable operation at every scale from IoT devices to multi-megawatt centers. If the mapping holds, practitioners gain a shared reference for comparing approaches and prioritizing research on the most power-hungry AI workloads.

Core claim

The paper establishes that current publications on energy-efficient computing can be systematically arranged into a holistic taxonomy built around hardware and software aspects, measurement instrumentation, software optimizations, dynamic task scheduling, voltage scaling, workload consolidation, federated learning, and cooling. This structure captures the state of the art for the full cloud-edge-HPC range and places special weight on the demands of large-scale generative AI training and inference. The resulting overview supports a forward roadmap for sustainable computing that accounts for financial, environmental, and supply constraints on energy.

What carries the argument

The holistic taxonomy that groups publications by perspectives including hardware/software aspects, measurement instrumentation, software optimizations, dynamic task scheduling, voltage scaling, workload consolidation, federated learning, and cooling.

If this is right

  • Researchers can use the taxonomy to locate under-explored intersections such as federated learning combined with dynamic voltage scaling.
  • System designers gain a shared language for comparing hardware and software energy strategies across IoT to HPC scales.
  • Roadmap priorities shift toward techniques that jointly address carbon footprint, energy supply limits, and large AI workloads.
  • Measurement and cooling methods become more visible as first-class components rather than afterthoughts in energy studies.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The taxonomy could be turned into a living database that automatically classifies new papers as they appear.
  • Applying the same structure to older literature might reveal whether today's dominant approaches differ in kind from earlier ones.
  • If cooling and federated learning remain separate branches, future work may need explicit cross-links when both are used on the same cluster.

Load-bearing premise

The publications chosen for review represent the most recent and significant contributions, so the taxonomy built from them accurately reflects the overall state of the field.

What would settle it

A major 2025 or 2026 paper on energy reduction in exascale AI or data centers whose methods cannot be placed in any of the taxonomy's listed categories without stretching the definitions.

Figures

Figures reproduced from arXiv: 2605.24569 by Claude Tadonki, Roblex Nana Tchakoute.

Figure 1
Figure 1. Figure 1: The whole cycle of energy consumption and related consequences. This paper aims to provide a structured and commented view of the landscape of the major scientific and techni￾cal contributions within the state-of-the-art of power-aware computing in the Computing continuum. We provide and articulate the following viewpoints: • Comprehensive Taxonomy: An up-to-date taxon￾omy of energy-related studies spannin… view at source ↗
Figure 2
Figure 2. Figure 2: The landscape of energy-aware computing in the Cloud-Edge-HPC continuum. 2.3. Overview of the Major Selected Contributions We made a specific methodological choice regarding the selection of the papers for our corpus as previously de￾scribed. We acknowledge that this choice may have resulted in the omission of genuine/valuable/valued contributions (less recent ones, books, blogs, conferences/journals with … view at source ↗
Figure 3
Figure 3. Figure 3: Panorama of the keywords from our corpus. R.N. Tchakoute and C. Tadonki: Preprint submitted to Elsevier Page 3 of 26 [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Distribution of recent hardware-related energy-aware studies across the computing continuum. 5. Energy and Carbon Profiling Tools Dynamic energy optimization strategies across the Cloud￾Edge-HPC continuum require real-time, fine-grained teleme￾try. As hardware diversity expands, the abstraction gap between software workflow and hardware profile is widen￾ing. The profiling ecosystem can be categorized into … view at source ↗
Figure 5
Figure 5. Figure 5: Overview of energy measurement methodologies. 6. Software-Level Energy Optimization While high-performance chips define minimum energy consumption, a system’s carbon footprint is largely deter￾mined by its software. Thus, this carbon footprint can only be mitigated through advanced and comprehensive soft￾ware optimization: from code compilation to operating sys￾tem power management, including distributed c… view at source ↗
Figure 6
Figure 6. Figure 6: Prevalence of software-level energy optimization strategies within recent literature. 7. The Energy Wall of Generative AI The impressive emergence of Generative Artificial In￾telligence (GenAI) has significantly magnified the level of resources needed in data centers. Since 2012, the computing resources needed to train advanced AI systems is doubling approximately every 3.4 months [245]. As of early 2026, … view at source ↗
Figure 7
Figure 7. Figure 7: Quantitative overview of Green-AI literature. R.N. Tchakoute and C. Tadonki: Preprint submitted to Elsevier Page 13 of 26 [PITH_FULL_IMAGE:figures/full_fig_p013_7.png] view at source ↗
read the original abstract

High-Performance Computing (HPC) has recently entered the Exascale era, and considerable efforts are being made to fully harness this potential power for large-scale applications, such as cutting-edge generative AI (training and exploitation). The corresponding energy consumption is very high, and forecasts are alarming, making this metric a critical systemic bottleneck. Addressing this issue presents a genuine challenge for the entire cloud-edge-HPC continuum at all scales, from low-power IoT microcontrollers to multi-megawatt data centers. Beyond financial costs, green computing is driven by considerations related to climate change and environmental concerns such as carbon footprint ($CO_2e$), as well as constraints on energy production and supply, leading to a real need to regulate {\em information and communication technology} (ICT) activities. This article presents a comprehensive overview of energy-efficient computing, taking into account the most recent and significant contributions. Based on this exploration of the state of the art, we design and describe a holistic taxonomy of the aforementioned publications, structured around various perspectives, including {\em hardware and software aspects, measurement instrumentation, software optimizations, dynamic task scheduling, voltage scaling, workload consolidation, federated learning}, and {\em cooling}. Particular emphasis is placed on large-scale AI, which receives significant attention due to its considerable resource requirements. We conclude with an analysis of a forward-looking roadmap that considers the main perspectives of sustainable computing.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 0 minor

Summary. The manuscript surveys energy-efficient computing across the cloud-edge-HPC-IoT continuum, with emphasis on large-scale AI workloads. It reviews recent contributions and constructs a holistic taxonomy organized by hardware/software aspects, measurement instrumentation, software optimizations, dynamic task scheduling, voltage scaling, workload consolidation, federated learning, and cooling, concluding with a forward-looking roadmap for sustainable computing.

Significance. A structured taxonomy of energy-aware methods could usefully organize an interdisciplinary field and surface connections among hardware, software, and system-level techniques, particularly for AI. The survey's value as a reference depends on the taxonomy's coverage; the absence of documented selection criteria reduces its reliability as a definitive state-of-the-art map.

major comments (1)
  1. [Abstract] Abstract: the central claim that a 'holistic taxonomy' can be constructed from 'the most recent and significant contributions' across the listed perspectives cannot be evaluated, because no section describes the literature search (databases, keywords, date range, screening process, or total papers considered). This omission is load-bearing for the taxonomy's asserted completeness.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the constructive feedback. We address the major comment below and will incorporate changes in the revised manuscript.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the central claim that a 'holistic taxonomy' can be constructed from 'the most recent and significant contributions' across the listed perspectives cannot be evaluated, because no section describes the literature search (databases, keywords, date range, screening process, or total papers considered). This omission is load-bearing for the taxonomy's asserted completeness.

    Authors: We agree that the absence of an explicit literature search methodology section limits the ability to assess the taxonomy's scope and completeness. In the revised version, we will add a dedicated subsection (likely in Section 1 or as a new Section 2) that documents the search process. This will include the academic databases queried (e.g., IEEE Xplore, ACM DL, arXiv, ScienceDirect), search keywords and Boolean strings, the date range (primarily 2020–2025 with selected foundational works), inclusion/exclusion criteria, and approximate totals of papers screened and retained. This addition will directly support the claim of a holistic taxonomy derived from recent contributions. revision: yes

Circularity Check

0 steps flagged

No circularity: survey taxonomy has no derivations or self-referential reductions

full rationale

This is a literature review paper that surveys energy-efficient computing publications and organizes them into a taxonomy across listed perspectives. No equations, fitted parameters, predictions, or derivations appear anywhere in the manuscript. The taxonomy is explicitly constructed from the reviewed works rather than claiming to derive new results from first principles or reduce to self-citation chains. Absence of a documented search protocol affects verifiability of completeness but does not create any of the enumerated circularity patterns (self-definitional, fitted-input prediction, load-bearing self-citation, etc.). The paper is self-contained as a descriptive overview.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

This is a survey paper; it introduces no free parameters, axioms, or invented entities.

pith-pipeline@v0.9.1-grok · 5778 in / 1000 out tokens · 20262 ms · 2026-06-30T12:17:58.963873+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

291 extracted references · 98 canonical work pages · 3 internal anchors

  1. [1]

    IEA dataset/report reference

    International Energy Agency, Co2 emissions associated with electricity generation for data centres,https: //www.iea.org/data-and-statistics/charts/co2-emissions-\ associated-with-electricity-generation-for-data-centres-by-case\ -2020-2035, 2025. IEA dataset/report reference

    2020

  2. [2]

    top500.org/lists/top500/, 2025

    TOP500 Organization, Top500 supercomputer sites,https://www. top500.org/lists/top500/, 2025. June/November 2025 edition

    2025

  3. [3]

    Official blog and technical overview

    Meta AI, Introducing meta llama 3,https://ai.meta.com/blog/ meta-llama-3/, 2024. Official blog and technical overview

    2024

  4. [4]

    NVIDIA, Nvidia blackwell platform arrives to power a new era of computing,https://nvidianews.nvidia.com/news/ nvidia-blackwell-platform-arrives-to-power-a-new-era-of-computing,

  5. [5]

    Official announcement (whitepaper-style content)

  6. [6]

    com/en/products/accelerators/instinct/mi300.html, 2024

    AMD, Amd instinct mi300 series accelerators,https://www.amd. com/en/products/accelerators/instinct/mi300.html, 2024. Product whitepaper/specifications

    2024

  7. [7]

    United Nations Framework Convention on Climate Change, Paris agreement,https://unfccc.int/process-and-meetings/ the-paris-agreement/the-paris-agreement, 2015

    2015

  8. [8]

    A.G.Philipo,H.Ning,D.S.Sarwatt,J.A.Mohamed,A.S.Yusufu, F.Shi,S.Urenje,J.Ding, Sustainableai:Emergingtrends,impacts, and future challenges, IEEE Trans. Sustain. Comput. 10 (2025) 1278–1291

    2025

  9. [9]

    W.Lin,J.Lin,Z.Peng,H.Huang,W.Lin,K.Li,Asystematicreview of green-aware management techniques for sustainable data center, Sustain. Comput. Inf. Syst. 42 (2024) 100989

    2024

  10. [10]

    S.M.Hasan,T.Islam,M.Saifuzzaman,K.R.Ahmed,C.-H.Huang, A. R. Shahid, Carbon emission quantification of machine learning: A review, IEEE Trans. Sustain. Comput. 10 (2025) 1085–1102

    2025

  11. [11]

    Representthemall: A universal learning representation of bug reports

    S.McGuire,E.Schultz,B.Ayoola,P.Ralph, Sustainabilityisstrati- fied:Towardabettertheoryofsustainablesoftwareengineering, in: IEEE/ACM45thInt.Conf.Soft.Eng.(ICSE),2023,pp.1996–2008. doi:10.1109/ICSE48619.2023.00169

    work page doi:10.1109/icse48619.2023.00169 2023

  12. [12]

    Accessed: 2026-04-16

    Google Cloud, Introducing trillium: Google’s 6th generation tpu,https://cloud.google.com/blog/products/compute/ introducing-trillium-6th-gen-tpus, 2024. Accessed: 2026-04-16

    2024

  13. [13]

    Parallel

    G.Raffin,D.Trystram, Dissectingthesoftware-basedmeasurement of cpu energy consumption: A comparative analysis, IEEE Trans. Parallel. Distrib. Syst. 36 (2025) 96–107

    2025

  14. [14]

    Z. Yang, K. Adamek, W. Armour, Accurate and convenient energy measurements for gpus: A detailed study of nvidia gpu’s built-in power sensor, in: SC24: Proc. Int. Conf. High Perform. Comput., Netw., Storage Anal, 2024, pp. 1–17. doi:10.1109/SC41406.2024. 00028

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1109/sc41406.2024 2024

  15. [15]

    Ilsche, D

    T. Ilsche, D. Hackenberg, S. Graul, R. Schöne, J. Schuchart, Power measurementsforcomputenodes:Improvingsamplingrates,granu- larity and accuracy, in: IEEE IGSC’15, 2015, pp. 1–8. doi:10.1109/ IGCC.2015.7393710

    work page arXiv 2015

  16. [16]

    S.Rajput,T.Widmayer,Z.Shang,M.Kechagia,F.Sarro,T.Sharma, Enhancing energy-awareness in deep learning through fine-grained energymeasurement,ACMTrans.Softw.Eng.Methodol.33(2024)

    2024

  17. [17]

    Guldner, R

    A. Guldner, R. Bender, C. Calero, G. S. Fernando, M. Funke, J. Gröger, L. M. Hilty, J. Hörnschemeyer, G.-D. Hoffmann, D.Junger,T.Kennes,S.Kreten,P.Lago,F.Mai,I.Malavolta,J.Mu- rach, K. Obergöker, B. Schmidt, A. Tarara, J. P. De Veaugh-Geiss, S. Weber, M. Westing, V. Wohlgemuth, S. Naumann, Development and evaluation of a reference measurement model for as...

    2024

  18. [18]

    Zhang, Z

    J. Zhang, Z. Wang, H. Wang, T. Song, H.-a. Su, R. Chen, Y. Hua, X. Zhou, R. Ma, M. Pan, H. Guan, Ampere: A generic energy R.N. Tchakoute and C. Tadonki:Preprint submitted to ElsevierPage 18 of 26 Energy-Aware Computing in 2026 estimationapproachforon-devicetraining, SIGMETRICSPerform. Eval. Rev. 53 (2025) 27–32

    2026

  19. [19]

    Z.Zong,R.Ge,Q.Gu,Marcher:Aheterogeneoussystemsupporting energy-aware high performance computing and big data analytics, BigDataResearch8(2017)27–38.TutorialsonToolsandMethods using High Performance Computing resources for Big Data

    2017

  20. [20]

    33rd ACM Int

    M.Weber,J.Dorn,S.Apel,N.Siegmund, Famlem,thefastmodular energy meter at code level, in: Proc. 33rd ACM Int. Conf. Found. Soft.Eng.,FSECompanion’25,ACM,2025,p.1129–1133.doi:10. 1145/3696630.3728595

    work page arXiv 2025

  21. [21]

    Treibig, G

    J. Treibig, G. Hager, G. Wellein, Likwid: A lightweight performance-orientedtoolsuiteforx86multicoreenvironments, in: 2010 39th International Conference on Parallel Processing Work- shops, 2010, pp. 207–216

    2010

  22. [22]

    IEEE/ACM 21st Int

    S.Rajput,M.Kechagia,F.Sarro,T.Sharma, Greenlight:Highlight- ing tensorflow apis energy footprint, in: Proc. IEEE/ACM 21st Int. Conf. Min. Softw. Repos. (MSR), 2024, pp. 304–308

    2024

  23. [23]

    SC ’24 Workshops, SC-W ’24, IEEE Press, 2025, p

    M. Paipuri, Ceems: A resource manager agnostic energy and emissionsmonitoringstack, in:Proc.SC’24Workshops,SC-W’24, IEEE Press, 2025, p. 1862–1866. doi:10.1109/SCW63240.2024.00233

    work page doi:10.1109/scw63240.2024.00233 2025

  24. [24]

    Terboven, R

    C. Terboven, R. Liem, J. Gracia, K. Haldar, J. Engels, P. Gies- selmann, D. Brayford, T. Wilde, C. Simmendinger, M. Marquardt, J. Eitzinger, T. Gruber, Ee-hpc a framework for energy efficient hpc system management, in: Proc. SC ’24 Workshops, 2024, pp. 1878–

    2024

  25. [25]

    doi:10.1109/SCW63240.2024.00236

    work page doi:10.1109/scw63240.2024.00236 2024

  26. [26]

    T. A. Rahmani, G. Belalem, S. A. Mahmoudi, O. R. Merad-Boudia, Real-time ai-powered monitoring for energy-efficient scheduling in multi-nodeheterogeneoussystems,FutureGener.Comput.Syst.181 (2026) 108428

    2026

  27. [27]

    K.A.Sanni,A.G.Andreou, Ahistoricalperspectiveonhardwareai inference, charge-based computational circuits and an 8 bit charge- based multiply-add core in 16 nm finfet cmos, IEEE J. Emerg. Sel Top. Circ. Syst. 9 (2019) 532–543

    2019

  28. [28]

    K.Wang,H.Zheng,Y.Li,A.Louri, Securenoc:Alearning-enabled, high-performance,energy-efficient,andsecureon-chipcommunica- tionframeworkdesign, IEEETrans.Sustain.Comput.7(2022)709– 723

    2022

  29. [29]

    M. H. Sadi, C. Sudarshan, S. R. Nassif, N. Wehn, Efficient deep neuralnetworktrainingwithanovel5.3-bitblockfloatingpointdata format, IEEE Trans. Circ Syst. A.I. 2 (2025) 150–161

    2025

  30. [30]

    Symp.High-Perform.Comput.Archit.(HPCA),2025,pp.409–422

    G.Li,S.Ye,C.Chen,Y.Wang,F.Yang,T.Cao,C.Liu,M.M.S.Aly, M. Yang, Lut-dla: Lookup table as efficient extreme low-bit deep learning accelerator, in: IEEE Int. Symp. High-Perform. Comput. Archit. (HPCA), 2025, pp. 671–684. doi:10.1109/HPCA61900.2025. 00057

    work page doi:10.1109/hpca61900.2025 2025

  31. [31]

    Laskar, P

    S. Laskar, P. Majhi, A. Muzahid, E. J. Kim, Supermesh: Energy- efficientcollectivecommunicationsforaccelerators, in:Proceedings ofthe58thIEEE/ACMInternationalSymposiumonMicroarchitec- ture, MICRO ’25, ACM, 2025, p. 1640–1655. doi:10.1145/3725843. 3756085

    work page doi:10.1145/3725843 2025

  32. [32]

    Y. Liu, Y. Xue, N. Crawford, J. Xue, J. Huang, Elk: Exploring the Efficiency of Inter-core Connected AI Chips with Deep Learning Compiler Techniques, ACM, 2025, p. 1284–1299

    2025

  33. [33]

    T. Li, H. Zhong, Y. Xu, V. Narayanan, K. Ni, H. Yang, T. Kämpfe, X. Li, Remna: Variation-resilient and energy-efficient mlc fefet computing-in-memory using nand flash-like read and adaptive con- trol, in: Proc. 43rd IEEE/ACM Int. Conf. Computer Aided Des. (ICCAD), ICCAD ’24, ACM, 2025. doi:10.1145/3676536.3676715

    work page doi:10.1145/3676536.3676715 2025

  34. [34]

    D. R. Dipta, B. Gulmezoglu, Mad-en: Microarchitectural attack detection through system-wide energy consumption, IEEE Trans. Inform. Forens. Secur. 18 (2023) 3006–3017

    2023

  35. [35]

    Ebrahimi, A

    Z. Ebrahimi, A. Kumar, Green: An approximate simd/mimd cgra forenergy-efficientprocessingattheedge, IEEETrans.Comp.-Aid. Des. Integr. Circ. Syst. 43 (2024) 2874–2887

    2024

  36. [36]

    F.Fayza,C.Demirkiran,S.P.Rao,D.Bunandar,U.Gupta,A.Joshi, Epicarbon: A carbon modeling tool for electro-photonic accelera- tors,in:2025IEEE/ACMInt.Conf.ComputerAidedDes.(ICCAD), 2025, pp. 1–9. doi:10.1109/ICCAD66269.2025.11240809

    work page doi:10.1109/iccad66269.2025.11240809 2025

  37. [37]

    Wu, Act: designing sustainable computer systems with an architectural carbon modeling tool, in: ACM/IEEE 49th Annu

    U.Gupta,M.Elgamal,G.Hills,G.-Y.Wei,H.-H.S.Lee,D.Brooks, C.-J. Wu, Act: designing sustainable computer systems with an architectural carbon modeling tool, in: ACM/IEEE 49th Annu. Int. Symp.Comput.Archit.(ISCA),ISCA’22,ACM,2022,p.784–799. doi:10.1145/3470496.3527408

    work page doi:10.1145/3470496.3527408 2022

  38. [38]

    Pashaeifar, M

    M. Pashaeifar, M. Kamal, A. Afzali-Kusha, M. Pedram, A the- oretical framework for quality estimation and optimization of dsp applicationsusinglow-powerapproximateadders, IEEETrans.Circ Syst. 66 (2019) 327–340

    2019

  39. [39]

    Danopoulos, G

    D. Danopoulos, G. Zervakis, D. Soudris, J. Henkel, Transaxx: Efficient transformers with approximate computing, IEEE Trans. Circ Syst. Art. Intell. 2 (2025) 288–301

    2025

  40. [40]

    Golden, M

    A. Golden, M. Elgamal, A. Mahmoud, G. Hills, C.-J. Wu, G.- Y. Wei, D. Brooks, Wafer-scale systems: A carbon perspective, SIGENERGY Energy Inform. Rev. 5 (2025) 118–124

    2025

  41. [41]

    Kalyanapu, F

    J. Kalyanapu, F. Dizani, D. Asher, A. Ghanbari, R. Cammarota, A.Aysu,S.M.Ajorpaz,GateBleed:ExploitingOn-CoreAccelerator Power Gating for High Performance and Stealthy Attacks on AI, ACM, 2025, p. 308–325

    2025

  42. [42]

    Abu Ahmad, A

    W. Abu Ahmad, A. Bartolini, F. Beneventi, L. Benini, A. Borghesi, M.Cicala,P.Forestieri,C.Gianfreda,D.Gregori,A.Libri,F.Spiga, S.Tinti, Designofanenergyawarepetaflopsclasshighperformance clusterbasedonpowerarchitecture,in:IEEEIPDPS’17Workshops, 2017, pp. 964–973. doi:10.1109/IPDPSW.2017.22

    work page doi:10.1109/ipdpsw.2017.22 2017

  43. [43]

    Splitwise: Efficient generative LLM inference using phase splitting,

    M. Cochet, K. Swaminathan, E. Loscalzo, J. Zuckerman, M. C. D. Santos, D. Giri, A. Buyuktosunoglu, T. Jia, D. Brooks, G.-Y. Wei, K. Shepard, L. P. Carloni, P. Bose, Blitzcoin: Fully decentral- ized hardware power management for accelerator-rich socs, in: ACM/IEEE 51st Annu. Int. Symp. Comput. Archit. (ISCA), 2024, pp. 801–817. doi:10.1109/ISCA59077.2024.00063

    work page doi:10.1109/isca59077.2024.00063 2024

  44. [44]

    A. M. Panteleaki, K. Balaskas, G. Zervakis, H. Amrouch, I. Anag- nostopoulos, Late breaking results: Leveraging approximate com- putingforcarbon-awarednnaccelerators, in:Des.Autom.TestEur. Conf. Exhib. (DATE), 2025, pp. 1–2. doi:10.23919/DATE64628.2025. 10993191

    work page doi:10.23919/date64628.2025 2025

  45. [45]

    C. C. Sudarshan, A. Arora, V. A. Chhabria, Greenfpga: Evaluating fpgas as environmentally sustainable computing solutions, in: 2024 61st ACM/IEEE Design Autom. Conf. (DAC), 2024, pp. 1–6

    2024

  46. [46]

    Darabi, E

    A. Darabi, E. Abiri, Cgi-sram: Memory cell with data-aware write bitline-free and inner readout features for energy-efficient bitwise logic-in-memory operations, Sustain. Comput. Inf. Syst. 50 (2026) 101307

    2026

  47. [47]

    J.Spieck,D.Walter,J.Waschkeit,J.Teich, Co-designofsustainable embedded systems-on-chip, in: Des. Autom. Test Eur. Conf. Exhib. (DATE), 2025, pp. 1–2. doi:10.23919/DATE64628.2025.10992914

    work page doi:10.23919/date64628.2025.10992914 2025

  48. [48]

    Shiflett, A

    K. Shiflett, A. Karanth, R. Bunescu, A. Louri, Albireo: Energy- efficient acceleration of convolutional neural networks via silicon photonics, in: ACM/IEEE 48th Annu. Int. Symp. Comput. Archit. (ISCA), 2021, pp. 860–873. doi:10.1109/ISCA52012.2021.00072

    work page doi:10.1109/isca52012.2021.00072 2021

  49. [49]

    J. Li, A. Louri, A. Karanth, R. Bunescu, Gcnax: A flexible and energy-efficientacceleratorforgraphconvolutionalneuralnetworks, in:IEEEInt. Symp.High-Perform.Comput.Archit. (HPCA),2021, pp. 775–788. doi:10.1109/HPCA51647.2021.00070

    work page doi:10.1109/hpca51647.2021.00070 2021

  50. [50]

    X. Li, T. Zhou, H. Wang, M. Lin, Energy-efficient computation with dvfs using deep reinforcement learning for multi-task systems in edge computing, IEEE Trans. Sustain. Comput. 10 (2025) 1116– 1127

    2025

  51. [51]

    A.Swaraj,S.Kumar, Ischatgpt-generatedcodereallygreen?:Eval- uating ai-generated solutions for energy-efficient coding practices, IEEE Software 43 (2026) 42–50

    2026

  52. [52]

    Georgiou, M

    S. Georgiou, M. Kechagia, T. Sharma, F. Sarro, Y. Zou, Green ai: do deep learning frameworks have different costs?, in: Proc. 44th Int. Conf. Soft. Eng., ICSE ’22, ACM, 2022, p. 1082–1094. doi:10.1145/3510003.3510221

    work page doi:10.1145/3510003.3510221 2022

  53. [53]

    Twin graph-based anomaly detection via attentive multi-modal learning for microservice system

    S. Sikand, V. S. Sharma, V. Kaulgud, S. Podder, Green ai quotient: Assessing greenness of ai-based software and the way forward, in: Proceedings of the 38th IEEE/ACM International Conference on Automated Software Engineering, ASE ’23, IEEE Press, 2024, p. R.N. Tchakoute and C. Tadonki:Preprint submitted to ElsevierPage 19 of 26 Energy-Aware Computing in 2...

    work page doi:10.1109/ase56229.2023.00115 2024

  54. [54]

    D.Cesarini,A.Bartolini,P.Bonfà,C.Cavazzoni,L.Benini, Count- down: A run-time library for performance-neutral energy saving in mpi applications, IEEE Trans. Comput. 70 (2021) 682–695

    2021

  55. [55]

    Stoico, A

    V. Stoico, A. C. Dragomir, P. Lago, An empirical study on the performance and energy usage of compiled python code, in: Proc. 29thInt.Conf.Eval.Assess.Softw.Eng.(EASE),EASE’25,ACM, 2025, p. 46–56. doi:10.1145/3756681.3756972

    work page doi:10.1145/3756681.3756972 2025

  56. [56]

    J. Song, P. Zhu, Y. Zhang, G. Yu, Cloudsimper: Simulating geo- distributed datacenters powered by renewable energy mix, IEEE Trans. Parallel. Distrib. Syst. 35 (2024) 531–547

    2024

  57. [57]

    Tunzina, M

    T. Tunzina, M. Mashira, M. M. Islam, Greenai: A comparative analysis of environmental efficiency in llm-generated code, IEEE Access 14 (2026) 18654–18672

    2026

  58. [58]

    De Martino, S

    V. De Martino, S. Martínez-Fernández, F. Palomba, Do developers adopt green architectural tactics for ml-enabled systems? a mining softwarerepositorystudy, in:IEEE/ACM47thInt.Conf.Soft.Eng. (ICSE-SEIS),2025,pp.135–139.doi:10.1109/ICSE-SEIS66351.2025. 00019

    work page doi:10.1109/icse-seis66351.2025 2025

  59. [59]

    N. R. Shah, M. V. V. S. M. Kumar, D. Baxi, R. Upadrasta, Polyufc: Polyhedralcompilationmeetsrooflineanalysisforuncorefrequency capping, in: 2026 IEEE/ACM International Symposium on Code Generation and Optimization (CGO), 2026, pp. 563–576. doi:10. 1109/CGO68049.2026.11395211

    work page arXiv 2026

  60. [60]

    M. Hort, A. Grishina, L. Moonen, An exploratory literature study on sharing and energy use of language models for source code, in: ACM/IEEE Int. Symp. Emp. Soft. Eng Meas. (ESEM), 2023, pp. 1–12. doi:10.1109/ESEM56168.2023.10304803

    work page doi:10.1109/esem56168.2023.10304803 2023

  61. [61]

    IEEE Press, 2025

    B. Ayoola, M. Kuutila, R. R. Wehbe, P. Ralph, User personas improvesocialsustainabilitybyencouragingsoftwaredevelopersto deprioritize antisocial features, in: IEEE/ACM 47th In. Conf. Soft. Eng.(ICSE),2025,pp.833–845.doi:10.1109/ICSE55347.2025.00135

    work page doi:10.1109/icse55347.2025.00135 2025

  62. [62]

    J. Chen, L. Tan, P. Wu, D. Tao, H. Li, X. Liang, S. Li, R. Ge, L.Bhuyan,Z.Chen, Greenla:Greenlinearalgebrasoftwareforgpu- accelerated heterogeneous computing, in: SC ’16: Proc. Int. Conf. High Perform. Comput., Netw., Storage Anal, 2016, pp. 667–677. doi:10.1109/SC.2016.56

    work page doi:10.1109/sc.2016.56 2016

  63. [63]

    N.v.Kempen,H.-J.Kwon,D.T.Nguyen,E.D.Berger, It’snoteasy beinggreen:Ontheenergyefficiencyofprogramminglanguages,in: 40th IEEE/ACM Int. Conf. Aut. Soft. Eng. (ASE), 2025, pp. 1553–

    2025

  64. [64]

    doi:10.1109/ASE63991.2025.00131

    work page doi:10.1109/ase63991.2025.00131 2025

  65. [65]

    L. Cruz, X. Franch, S. Martínez-Fernández, Innovating for tomor- row: The convergence of software engineering and green ai, ACM Trans. Softw. Eng. Methodol. 34 (2025)

    2025

  66. [66]

    De Martino, S

    V. De Martino, S. Lambiase, F. Pecorelli, W.-J. van den Heuvel, F.Ferrucci,F.Palomba, Sustainabilityofmachinelearning-enabled systems: The machine learning practitioner’s perspective, ACM Trans. Softw. Eng. Methodol. (2025). Just Accepted

    2025

  67. [67]

    2406–2418

    E.B.Roque,L.Cruz,T.Durieux,UnveilingtheEnergyVampires:A Methodology for Debugging Software Energy Consumption, IEEE Press, 2025, p. 2406–2418

    2025

  68. [68]

    Maquoi, M

    J. Maquoi, M. Cauz, B. Vanderose, X. Devroey, Energy codesump- tion, leveraging test execution for source code energy consumption analysis, in: Proc.33rd ACM Int. Conf. Found. Soft. Eng., FSE Companion ’25, ACM, 2025, p. 1432–1436. doi:10.1145/3696630. 3728707

    work page doi:10.1145/3696630 2025

  69. [69]

    IEEE/ACM 22nd Int

    N.Alizadeh,B.Belchev,N.Saurabh,P.Kelbert,F.Castor,Language models in software development tasks: An experimental analysis of energy and accuracy, in: Proc. IEEE/ACM 22nd Int. Conf. Min. Softw. Repos. (MSR), IEEE Press, 2025, p. 725–736. doi:10.1109/ MSR66628.2025.00109

    work page arXiv 2025

  70. [70]

    Zanfardino, M

    G. Zanfardino, M. A. Memon, M. Tucci, Enhancing energy effi- ciency with reusable software ecosystems and persona-based ui/ux, in: Proc.33rd ACM Int. Conf. Found. Soft. Eng., FSE Companion ’25, ACM, 2025, p. 1449–1452. doi:10.1145/3696630.3728710

    work page doi:10.1145/3696630.3728710 2025

  71. [71]

    A.Tundo,M.Mobilio,S.Ilager,I.Brandić,E.Bartocci,L.Mariani, An energy-aware approach to design self-adaptive ai-based applica- tions on the edge, in: Proceedings of the 38th IEEE/ACM Interna- tional Conference on Automated Software Engineering, ASE ’23, IEEE Press, 2024, p. 281–293. doi:10.1109/ASE56229.2023.00046

    work page doi:10.1109/ase56229.2023.00046 2024

  72. [72]

    Andrés Larracoechea, P

    J. Andrés Larracoechea, P. Roose, S. Ilarri, Energy-aware software design: A hardware and behavioural consumption scoring and la- belling algorithm, Expert Systems with Applications 299 (2026) 130046

    2026

  73. [73]

    Dhawan, M

    R. Dhawan, M. Suresh Patel, Carbon-aware ai control plane for devops automation: A reference architecture and next-generation sustainability framework, IEEE Access 14 (2026) 11163–11184

    2026

  74. [74]

    Marahatta, Q

    A. Marahatta, Q. Xin, C. Chi, F. Zhang, Z. Liu, Pefs: Ai-driven prediction based energy-aware fault-tolerant scheduling scheme for clouddatacenter, IEEETrans.Sustain.Comput.6(2021)655–666

    2021

  75. [75]

    J. S. Shim, B. Han, Y. Kim, J. Kim, Deeppm: Transformer-based power and performance prediction for energy-aware software, in: Des. Autom. Test Eur. Conf. Exhib. (DATE), 2022, pp. 1491–1496. doi:10.23919/DATE54114.2022.9774589

    work page doi:10.23919/date54114.2022.9774589 2022

  76. [76]

    M. A. N. Al-hayanni, A. Rafiev, F. Xia, R. Shafik, A. Romanovsky, A.Yakovlev, Parma:Parallelization-awarerun-timemanagementfor energy-efficientmany-coresystems,IEEETrans.Comput.69(2020) 1507–1518

    2020

  77. [77]

    L.Eeckhout, Assessingprocessorsustainabilityusingthefirst-order focal carbon model, IEEE Micro 45 (2025) 11–18

    2025

  78. [78]

    Z. Zhou, M. Shojafar, J. Abawajy, H. Yin, H. Lu, Ecms: An edge intelligent energy efficient model in mobile edge computing, IEEE Trans. Green Comm. Network. 6 (2022) 238–247

    2022

  79. [79]

    Conficoni, A

    C. Conficoni, A. Bartolini, A. Tilli, C. Cavazzoni, L. Benini, Inte- grated energy-aware management of supercomputer hybrid cooling systems, IEEE Trans. Indus. Inform. 12 (2016) 1299–1311

    2016

  80. [80]

    Vahabi, F

    S. Vahabi, F. Righetti, C. Vallati, N. Tonellotto, The impact of prediction models on energy-aware resource management in faas platforms, IEEE Access 13 (2025) 85711–85727

    2025

Showing first 80 references.