Pith. sign in

REVIEW 4 major objections 5 minor 1 cited by

Carbon-Aware Temporal Data Transfer Scheduling Across Cloud Datacenters

T0 review · 4 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read LinTS, a linear-programming scheduler, cuts inter-datacenter transfer carbon by up to 66% versus worst case and up to 15% versus heuristics, while meeting all deadlines.

desk verdict A plausible carbon-aware transfer scheduler whose LP objective doesn't match its own simulator's emission metric; worth a major-revision referee, not a desk reject. read the letter →

arxiv 2506.04117 v1 pith:7BQ7KHZ7 submitted 2025-06-04 cs.DC cs.NI

Elvis Rodrigues , Jacob Goldverg , Tevfik Kosar This is my paper
classification cs.DCcs.NI
keywords carbon-awareschedulingtemporalshiftingdatatransfersclouddatacenterslinearprogrammingcarbonintensitythreadscalingdeadlineconstraints
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper tries to establish that the carbon cost of moving data between cloud datacenters can be cut substantially, without missing deadlines, by choosing both when each transfer runs and how many threads it uses. Its scheduler, LinTS, casts the problem as a linear program that minimizes carbon-intensity-weighted throughput over time slots, subject to per-request completion, deadline, and bandwidth constraints. In simulations with 2024 US grid carbon traces and 200 transfers, LinTS beats first-come-first-serve and threshold-based heuristics by roughly 10-15% and the constructed worst case by up to 66%, while meeting every deadline. The result matters because inter-datacenter traffic is large, growing, and largely delay-tolerant, so temporal shifting is an available lever for cutting cloud emissions.

What carries the argument

The load-bearing object is the linear program itself, built on a linearized power model $P(\rho)=\frac{\Delta P}{L}\rho+P_{\min}$ that ties CPU power draw to transfer throughput $\rho$. The LP encodes deadlines through the dimension of the throughput vector, completion through byte-sum constraints, and link sharing through per-slot bandwidth constraints; its objective is the carbon-weighted throughput sum. A separate throughput-to-threads model $\rho(\theta)=L(1-\frac{1}{s_\rho L\theta+1})$ and its inverse let LinTS turn the continuous LP solution into a concrete thread-scaling plan.

What would settle it

Run LinTS's schedule and a naive earliest-deadline schedule on the same real transfer path, with the same files and deadlines, while metering total node power; if the measured energy difference is much smaller than the 10-15% that the simulator predicts, the linear power model is what failed.

Watch

Extended reading notes

Core claim

LinTS's central claim is that a linear-programming scheduler can produce lower-carbon transfer plans than the heuristics used in current transfer services. Each request has a file size and a hard deadline; the path's carbon intensity per 15-minute slot is the sum of regional intensities along the path. The LP minimizes $\sum_{i,j} c_{i,j}\rho_{i,j}$, the carbon-weighted throughput, under constraints that each request's bytes complete by its deadline, that total allocated bandwidth in any slot stays under the bottleneck $L$, and that each request's throughput is between 0 and $L$. The solver returns a throughput plan, which LinTS converts into a thread plan using the inverse of its throughput-versus-threads model. Over 200 simulated requests of 10-50 GB with 48-71 hour deadlines on an 8-node path, with 5% and 15% noise added to Electricity Maps traces, the paper reports up to 66% lower emissions than the worst case and up to 15% lower than the comparison heuristics while satisfying all deadlines.

Load-bearing premise

The whole saving rests on the assumption that every node's CPU power draw rises linearly with transfer throughput, so that running slower in greener hours actually burns less carbon; if real network equipment draws mostly fixed power regardless of throughput, the predicted savings would shrink or disappear.

Editorial extensions

If this is right

  • Providers can deploy LinTS as a library or REST endpoint for delay-tolerant replication and backup traffic, cutting transfer carbon by roughly 10-15% without changing file sizes or deadlines.
  • Thread scaling adds a second lever: because power is modeled as rising with throughput, the scheduler slows transfers in dirty hours and speeds them up in clean hours, rather than always running at maximum threads.
  • The LP formulation is extensible: adding constraints would enable spatiotemporal scheduling, and multi-objective solvers could trade off carbon against cost or latency, as the paper notes.
  • The size of the saving depends on how much carbon intensity varies over time; with flatter or greener grids the scheduling advantage narrows.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Our inference: the LP minimizes a proxy that omits the constant idle-power term $P_{\min}$; if real nodes draw mostly fixed power regardless of throughput, the true energy gap between schedules will be smaller than simulated. Re-running the evaluation with a stepped or state-based power model would bound this effect.
  • Our inference: the simulator charges no energy to slots with no threads, but routers and switches along a path consume power continuously; adding per-node idle power would reduce the absolute savings even if the ranking of schedules stays the same.
  • Our inference: combining LinTS's temporal decisions with spatial replica placement (choosing a greener source region) could compound the savings; the paper does not evaluate that combination.
  • Our inference: the comparison baseline is a deliberately constructed worst case, so the 66% figure is an upper bound on improvement; benchmarking against production transfer logs with realistic arrival patterns would show how the savings generalize.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. The paper introduces LinTS, a linear-programming-based scheduler for inter-datacenter data transfers. LinTS chooses per-slot throughput (converted to thread counts) to minimize the carbon-weighted throughput over a job's allowed slots, subject to a byte-completion deadline, a per-slot bandwidth cap, and throughput bounds. The authors evaluate LinTS in simulation against FCFS, EDF, single-threshold, double-threshold, and worst-case schedules, using Electricity Maps carbon-intensity traces and a fitted CPU power model, and report carbon savings of up to 66% over the worst case and about 10-15% over the heuristics.

Significance. The underlying idea is timely and practical: inter-datacenter transfers are delay-tolerant, and temporal carbon-intensity variation can be exploited with a lightweight optimizer that also makes thread-scaling decisions. If the optimization objective were faithful to the emissions metric, the 10-15% improvement over common heuristics would be a useful contribution. The machine-checkable LP formulation and the inclusion of a real carbon-intensity trace evaluation are also positive features. However, the central quantitative claim is undermined by an objective mismatch between the LP and the simulator's emissions calculation, so the reported savings cannot currently be taken at face value.

major comments (4)
  1. [Section III-A, Eq. (8), and Section III-B LP] The LP objective omits the per-active-slot baseline power Pmin, so LinTS does not minimize the emissions metric that the simulator computes. Equation (7) gives P(rho)=DeltaP/L*rho+Pmin, and the actual carbon cost of a job using active slots is sum_j c_j (DeltaP/L*rho_j + Pmin) = (DeltaP/L) sum_j c_j rho_j + Pmin * sum_j c_j. The LP in Section III-B minimizes only the first term. Because the deadline constraint J_i <= sum_j t_i,j rho_i,j is an inequality and the number of active slots is not otherwise penalized, the LP can spread a transfer over many low-carbon slots at low throughput; the simulator then charges Pmin for each such slot. With Pmin=88W and DeltaP=12W, the omitted baseline term is about 88% of the per-slot power, so it dominates. The paper's claim that linear constraints account for the cost of slower transfers is incorrect: those constraints guarantee only that enough bytes are transferred by the deadline, not that the number of active slots is minimized. Thus the optimization objective in Section III-B is not the evaluation metric in Tables II and III, and the reported savings may be an artifact of this mismatch.
  2. [Section III-A, Eqs. (4)-(7)] The linearization from Eq. (6) to Eq. (7) is asserted without an error bound, and Eq. (4) appears to contain a typo. The inverse of the throughput model in Eq. (1) is theta(rho) = rho/(s_rho L (L-rho)), not the expression with s_P given in Eq. (4). With the stated parameters (s_rho=1/24, s_P=1/50, DeltaP=12W, L=0.25/0.5/0.75 Gbps), the constant K in Eq. (5) is not 1, so P(rho)=DeltaP/L*rho+Pmin is not exact. The LP objective uses the linearized form while the simulator evaluates emissions from the nonlinear Eq. (3); the authors need either an approximation guarantee or a demonstration that the error is negligible over the feasible range for the actual parameters.
  3. [Section III-C, Algorithm 1, line 20] The line 'bub <- -8 * data size vec' is unexplained. The slot length is 15 minutes (72 hours divided into 288 slots) and the data sizes are in GB, so the hard-coded constant 8 does not obviously follow from any defined unit conversion. This constant directly changes the byte-completion constraint and therefore affects every plan generated by LinTS. It must be documented and justified, or the algorithm pseudocode and the reproducible implementation must be corrected.
  4. [Section IV-A and Section IV-B] The evaluation is self-consistent rather than validated: the LP's power model and the simulator's emissions calculation both use Eq. (3) with parameters fitted on a single Chameleon Cloud path, and the authors assume all nodes along a multi-hop path have the same power behavior. Because the heuristics are also evaluated on the same model, the ranking may be a property of the model rather than of real transfer energy consumption. A sensitivity analysis varying Pmin, DeltaP, s_rho, s_P, and the functional form of P(rho), plus a real-world transfer experiment measuring actual CPU power and carbon intensity across a path, would be needed to support the quantitative claims in the abstract and Section IV-B.
minor comments (5)
  1. [Eq. (4)] Equation (4) should use s_rho in place of s_P when inverting Eq. (1); as written, the two equations are not inverses of each other.
  2. [Table II caption] The caption says '25%, 50%, and 40% of the first-hop bandwidth' but the text and Table III indicate the third level is 75%.
  3. [Section III-A, Eq. (8)] The vector p in Eq. (8) is not defined, and pmin is written as a scalar where the omitted term should be Pmin times a sum over active slots; this notation contributed to the objective mismatch discussed above.
  4. [Section III-C, Algorithm 1, line 8] The loop bound 'for n from 0 to num jobs' is ambiguous: if exactly num jobs iterations are intended, the upper bound should be num_jobs-1; otherwise the byte constraint may have an off-by-one error.
  5. [Section IV-C] The background-traffic experiment in Figure 4 is only descriptive; the paper does not quantify how the measured throughput variability would degrade LinTS schedules, despite identifying this as a limitation.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity found; the shared power model is an evaluation assumption, not a fitted prediction.

full rationale

The paper's derivation chain does not reduce to its own inputs in any of the enumerated ways. LinTS's LP objective is derived from a linearized power-throughput model (Eq 7), and the simulator evaluates plans with the same underlying power model (Eq 3); this shared model is an explicit evaluation assumption, not a parameter fitted to the outcome being claimed. The paper does not claim to validate the power model against real transfer emissions; it claims only that, under the stated model and 2024 Electricity Maps traces, LinTS's plans have lower simulated emissions than the heuristic baselines. The heuristics do not use the LP objective, so LinTS's advantage is not forced by construction. The self-citations (refs 37, 38, 39, 45) are background and motivational, and the linear-model assumption is justified by the paper's own Chameleon Cloud measurements, so no load-bearing result rests on an unverified self-citation. A real correctness concern exists at Section III-A (Eq 8) and the Section IV-A Simulator paragraph, but it is not circular: Eq 8 writes the per-slot Pmin term as a constant pmin, while the number of active slots is a free decision variable, so the LP objective is not identical to the emissions metric computed from Eq 3; this mismatch may affect the reported savings but makes the result an in-simulation empirical question rather than a tautology.

Assumptions & free parameters 5 free parameters · 6 assumptions · 0 invented entities

The central claim rests on a small set of fitted constants and strong modeling assumptions. The throughput and power scale constants are fit to a single experimental path, and the same model is used both to construct schedules and to evaluate them. The equal-weight path intensity and the CPU-only power model are assertions without direct evidence. No new physical entities are introduced.

free parameters (5)
  • Throughput scale constant s_rho = 1/24
    Set from a Chameleon Cloud experiment measuring throughput versus thread count (Section III.A); used in the inverse throughput-to-thread mapping and in the simulator.
  • Power scale constant s_P = 1/50
    Set from the same experiment measuring CPU power versus thread count; appears in Eq 3 and the linearized power model.
  • CPU power range Pmax and Pmin = 100 W and 88 W
    Assumed for the transfer node in Section IV.A; no reference or measurement is provided, and the values are applied to every node in the path.
  • Path node weights = Equal weights
    The simulator assigns equal weight to all nodes on a transfer path when summing carbon intensity traces (Section IV.A), without empirical justification.
  • Forecast noise levels = 5% and 15%
    Hand-picked to model forecast error (Section IV.A); no sensitivity analysis is reported across the range of plausible errors.
assumptions (6)
  • domain assumption The CPU power of a transfer is a linear function of throughput: P(rho) = (DeltaP/L) * rho + Pmin (Eq 7).
    This linearization is the basis of the LP objective and is asserted without quantifying the error relative to the nonlinear model in Eq 6 (Section III.A).
  • domain assumption All transfer requests are delay-tolerant, interruptible, and schedulable.
    Stated in Section III.A; real transfers may incur penalties for interruption or may be latency-sensitive.
  • domain assumption All nodes along a transfer path are equally affected by the transfer, so their carbon intensities can be summed with equal weight.
    Assumed in Section IV.A; no evidence is given that intermediate network equipment consumes power in proportion to transfer throughput.
  • domain assumption CPU power is the only power that varies with transfer throughput; network equipment and other components are ignored or implicitly included.
    The model uses only CPU power (Pmax/Pmin, Section IV.A), so the reported emissions may not reflect the full carbon cost of a transfer.
  • domain assumption Carbon intensity forecasts are available with bounded error (5-15%) and LinTS uses them without online re-planning.
    LinTS does not adjust plans as grid or network conditions change (Sections III.C and IV.A), limiting real-world applicability.
  • standard math The LP solver returns a globally optimal solution to the stated linear program.
    Assumed by using SciPy's linprog (Section III.C).

how reviews work

0 comments
Cite this review

Pith. "Pith review of Carbon-Aware Temporal Data Transfer Scheduling Across Cloud Datacenters." pith.science (2026). https://pith.science/paper/7BQ7KHZ7

@misc{pith2026250604117,
  author       = {Pith},
  title        = {Pith review of: Carbon-Aware Temporal Data Transfer Scheduling Across Cloud Datacenters},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/7BQ7KHZ7}},
  note         = {Machine review of arXiv:2506.04117}
}
read the original abstract

Inter-datacenter communication is a significant part of cloud operations and produces a substantial amount of carbon emissions for cloud data centers, where the environmental impact has already been a pressing issue. In this paper, we present a novel carbon-aware temporal data transfer scheduling framework, called LinTS, which promises to significantly reduce the carbon emission of data transfers between cloud data centers. LinTS produces a competitive transfer schedule and makes scaling decisions, outperforming common heuristic algorithms. LinTS can lower carbon emissions during inter-datacenter transfers by up to 66% compared to the worst case and up to 15% compared to other solutions while preserving all deadline constraints.

Figures

Figures reproduced from arXiv: 2506.04117 by the authors.

Figure 1
Figure 1. Spatial and temporal variations in carbon intensity. [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Modeling the correlation between power demand and achieved throughput. [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Comparison of algorithms’ carbon emissions when restricted to 25%, 50%, and 75% of first-hop capacity. [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗

Discussion (0). Sign in to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. A Survey on Task Scheduling in Carbon-Aware Container Orchestration

    cs.SE 2025-08 conditional novelty 4.0 of 10

    A systematic survey categorizing carbon-aware Kubernetes scheduling algorithms along hardware/software and energy/carbon axes, with a proposed taxonomy.

Reference graph

Works this paper leans on

51 extracted references · 45 canonical work pages · cited by 1 Pith paper

  1. [1]

    Generational Growth: AI data centers and the coming US power surge,

    C. Davenport, B. Singer, N. Mehta, B. Lee, and J. Mackay, “Generational Growth: AI data centers and the coming US power surge,” Goldman Sachs, 04 2024. [Online]. Available: https://www.goldmansachs.com/pdfs/insights/pages/ generational-growth-ai-data-centers-and-the-coming-us-power-surge/ report.pdf

    work page 2024

  2. [2]

    Google Environmental Report 2024,

    “Google Environmental Report 2024,” 2024. [Online]. Available: https://sustainability.google/reports/google-2024-environmental-report

    work page 2024

  3. [3]

    Meta 2024 Sustainability Report,

    “Meta 2024 Sustainability Report,” 2024. [Online]. Available: https: //sustainability.atmeta.com/2024-sustainability-report

    work page 2024

  4. [4]

    2024 Environmental Sustainability Report,

    “2024 Environmental Sustainability Report,” 2024. [Online]. Available: https://www.microsoft.com/en-us/corporate-responsibility/ sustainability/report

    work page 2024

  5. [5]

    New nuclear clean energy agreement with Kairos Power,

    M. Terrell, “New nuclear clean energy agreement with Kairos Power,” Oct. 2024. [Online]. Available: https://blog.google/outreach-initiatives/ sustainability/google-kairos-power-nuclear-energy-agreement

    work page 2024

  6. [6]

    Carbon Explorer: A Holistic Framework for Designing Carbon Aware Datacenters,

    B. Acun, B. Lee, F. Kazhamiaka, K. Maeng, U. Gupta, M. Chakkar- avarthy, D. Brooks, and C.-J. Wu, “Carbon Explorer: A Holistic Framework for Designing Carbon Aware Datacenters,” in Proceedings of the 28th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, V olume 2 , ser. ASPLOS 2023. New York, NY , USA:...

    work page 2023

  7. [7]

    On the Limitations of Carbon-Aware Temporal and Spatial Workload Shifting in the Cloud,

    T. Sukprasert, A. Souza, N. Bashir, D. Irwin, and P. Shenoy, “On the Limitations of Carbon-Aware Temporal and Spatial Workload Shifting in the Cloud,” in Proceedings EuroSys’24 . New York, NY , USA: Association for Computing Machinery, 2024, p. 924–941

    work page 2024

  8. [8]

    Metrics for Sustainability in Data Centers,

    A. Gandhi, K. Ghose, K. Gopalan, S. R. Hussain, D. Lee, David, Liu, Z. Liu, P. McDaniel, S. Mu, and E. Zadok, “Metrics for Sustainability in Data Centers,” ACM SIGENERGY Energy Informatics Review , vol. 3, pp. 40–46, 2023. [Online]. Available: https://api.semanticscholar.org/CorpusID:251304999

    work page 2023

Show all 51 references
  1. [9]

    MOSAIC: A Multi- Objective Optimization Framework for Sustainable Datacenter Manage- ment,

    S. Qi, D. Milojicic, C. Bash, and S. Pasricha, “MOSAIC: A Multi- Objective Optimization Framework for Sustainable Datacenter Manage- ment,” 2023

    2023

  2. [10]

    Tree- house: A Case For Carbon-Aware Datacenter Software,

    T. Anderson, A. Belay, M. Chowdhury, A. Cidon, and I. Zhang, “Tree- house: A Case For Carbon-Aware Datacenter Software,” SIGENERGY Energy Inform. Rev. , vol. 3, no. 3, p. 64–70, 10 2023

    2023

  3. [11]

    Carbon-Aware Computing for Datacenters,

    A. Radovanovi ´c, R. Koningstein, I. Schneider, B. Chen, A. Duarte, B. Roy, D. Xiao, M. Haridasan, P. Hung, N. Care, S. Talukdar, E. Mullen, K. Smith, M. Cottman, and W. Cirne, “Carbon-Aware Computing for Datacenters,” IEEE Transactions on Power Systems , vol. 38, no. 2, p. 12...

    2023

  4. [12]

    Cisco Global Cloud Index 2015–2020

    J. Thomas Barnett, A. Sumits, S. Jain, U. Andra, and T. Khurana, “Cisco Global Cloud Index 2015–2020.” [Online]. Available: https://www.cisco.com/c/dam/m/en us/service-provider/ ciscoknowledgenetwork/files/622 11 15-16-Cisco GCI CKN 2015-2020 AMER EMEAR NOV2016.pdf

    2015

  5. [13]

    Global Electricity Review 2023 — Ember — ember-energy.org,

    M. Wiatros-Motyka, D. Jones, H. Broadbent, and N. Fulghum, “Global Electricity Review 2023 — Ember — ember-energy.org,” https: //ember-energy.org/latest-insights/global-electricity-review-2023/, 2023, [Accessed 30-03-2025]

    2023

  6. [14]

    Estimating the carbon footprint of the grand project, a multi-decade astrophysics experiment,

    C. Aujoux, K. Kotera, and O. Blanchard, “Estimating the carbon footprint of the grand project, a multi-decade astrophysics experiment,” Astroparticle Physics, vol. 131, p. 102587, 2021

    2021

  7. [15]

    Electricity Maps,

    “Electricity Maps,” 2024. [Online]. Available: https://www. electricitymaps.com

    2024

  8. [16]

    WattTime,

    “WattTime,” 2024. [Online]. Available: https://watttime.org

    2024

  9. [17]

    Carbon- Scaler: Leveraging Cloud Workload Elasticity for Optimizing Carbon- Efficiency,

    W. A. Hanafy, Q. Liang, N. Bashir, D. Irwin, and P. Shenoy, “Carbon- Scaler: Leveraging Cloud Workload Elasticity for Optimizing Carbon- Efficiency,” Proc. ACM Meas. Anal. Comput. Syst. , vol. 7, no. 3, 12 2023

    2023

  10. [18]

    Bds+: An inter-datacenter data replication system with dynamic bandwidth separation,

    Y . Zhang, X. Nie, J. Jiang, W. Wang, K. Xu, Y . Zhao, M. J. Reed, K. Chen, H. Wang, and G. Yao, “Bds+: An inter-datacenter data replication system with dynamic bandwidth separation,” IEEE/ACM Transactions on Networking , vol. 29, no. 2, pp. 918–934, 2021

    2021

  11. [19]

    Cadre: Carbon-aware data replication for geo-diverse services,

    Z. Xu, N. Deng, C. Stewart, and X. Wang, “Cadre: Carbon-aware data replication for geo-diverse services,” in 2015 IEEE International Conference on Autonomic Computing , 2015, pp. 177–186

    2015

  12. [20]

    Carbon-aware and fault- tolerant migration of deep learning workloads in the geo-distributed cloud,

    J. Park, D. Kim, J. Kim, J. Han, and S. Chun, “Carbon-aware and fault- tolerant migration of deep learning workloads in the geo-distributed cloud,” in 2024 IEEE 17th International Conference on Cloud Com- puting (CLOUD) , 2024, pp. 494–501

    2024

  13. [21]

    CarbonClipper: Optimal Algorithms for Carbon-Aware Spatiotemporal Workload Management,

    A. Lechowicz, N. H. Christianson, B. Sun, N. Bashir, M. Hajiesmaili, A. Wierman, and P. Shenoy, “CarbonClipper: Optimal Algorithms for Carbon-Aware Spatiotemporal Workload Management,” 2024. [Online]. Available: https://api.semanticscholar.org/CorpusID:271874702

    2024

  14. [22]

    Carbon Contain- ers: A System-level Facility for Managing Application-level Carbon Emissions,

    J. Thiede, N. Bashir, D. Irwin, and P. Shenoy, “Carbon Contain- ers: A System-level Facility for Managing Application-level Carbon Emissions,” in Proceedings of the 2023 ACM Symposium on Cloud Computing, ser. SoCC ’23. New York, NY , USA: Association for Computing Machinery, 2...

    2023

  15. [23]

    Carbon-aware computing for data centers with probabilistic performance guarantees,

    S. Hall, F. Micheli, G. Belgioioso, A. Radovanovi ´c, and F. D ¨orfler, “Carbon-aware computing for data centers with probabilistic performance guarantees,” 2024. [Online]. Available: https://arxiv.org/abs/2410.21510

    2024

  16. [24]

    Cucumber: Renewable-aware admission control for delay-tolerant cloud and edge workloads,

    P. Wiesner, D. Scheinert, T. Wittkopp, L. Thamsen, and O. Kao, “Cucumber: Renewable-aware admission control for delay-tolerant cloud and edge workloads,” in European Conference on Parallel Processing ,

  17. [25]

    CASA: A Framework for SLO and Carbon-Aware Autoscaling and Scheduling in Serverless Cloud Computing,

    S. Qi, H. Moore, N. Hogade, D. Milojicic, C. Bash, and S. Pasricha, “CASA: A Framework for SLO and Carbon-Aware Autoscaling and Scheduling in Serverless Cloud Computing,” ArXiv, vol. abs/2409.00550, 2024. [Online]. Available: https://arxiv.org/abs/ 2409.00550

    2024 arXiv

  18. [26]

    Let’s wait awhile: how temporal workload shifting can reduce carbon emissions in the cloud,

    P. Wiesner, I. Behnke, D. Scheinert, K. Gontarska, and L. Thamsen, “Let’s wait awhile: how temporal workload shifting can reduce carbon emissions in the cloud,” Proceedings of the 22nd International Middleware Conference , 2021. [Online]. Available: https://doi.org/10. 1145/34...

    2021

  19. [27]

    Real-time Carbon Footprint Minimization in Sustainable Data Centers with Reinforcement Learning,

    S. Sarkar, A. Naug, R. L. Gutierrez, A. Guillen, V . Gundecha, A. R. Babu, and C. Bash, “Real-time Carbon Footprint Minimization in Sustainable Data Centers with Reinforcement Learning,” in NeurIPS 2023 Workshop on Tackling Climate Change with Machine Learning , 2023

    2023

  20. [28]

    Low-carbon operation of data centers with joint workload sharing and carbon allowance trading,

    D. Yan, M.-Y . Chow, and Y . Chen, “Low-carbon operation of data centers with joint workload sharing and carbon allowance trading,” IEEE Transactions on Cloud Computing , vol. 12, pp. 750–761, 2024. [Online]. Available: https://doi.org/10.1109/TCC.2024.3396476

    2024

  21. [29]

    perf: Linux profiling with performance counters — perfwiki.github.io,

    “perf: Linux profiling with performance counters — perfwiki.github.io,” https://perfwiki.github.io/main/, [Accessed 28-03-2025]

    2025

  22. [30]

    Intel® 64 and IA-32 Architectures Software Developer Manuals — intel.com,

    “Intel® 64 and IA-32 Architectures Software Developer Manuals — intel.com,” https://www.intel.com/content/www/us/en/developer/articles/ technical/intel-sdm.html, [Accessed 25-03-2025]

    2025

  23. [31]

    [Online]

    NVML API Reference Guide , Nvidia, 2025. [Online]. Available: https://docs.nvidia.com/deploy/pdf/NVML API Reference Guide.pdf

    2025

  24. [32]

    Enhancing energy-awareness in deep learning through fine-grained energy measurement,

    S. Rajput, T. Widmayer, Z. Shang, M. Kechagia, F. Sarro, and T. Sharma, “Enhancing energy-awareness in deep learning through fine-grained energy measurement,” ACM Trans. Softw. Eng. Methodol. , vol. 33, no. 8, Dec. 2024. [Online]. Available: https://doi.org/10.1145/3680470

    2024 doi

  25. [33]

    GitHub - green-kernel/powerletrics: Powermetrics for Linux — github.com,

    “GitHub - green-kernel/powerletrics: Powermetrics for Linux — github.com,” https://github.com/green-kernel/powerletrics, Green Ker- nel, [Accessed 28-03-2025]

    2025

  26. [34]

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

    J. Wang, U. Gupta, and S. Akshitha, “Peeling Back the Carbon Curtain: Carbon Optimization Challenges in Cloud Computing,” Proceedings of the 2nd Workshop on Sustainable Computer Systems , 2023. [Online]. Available: https://api.semanticscholar.org/CorpusID:260380570

    2023

  27. [35]

    Going Green for Less Green: Optimizing the Cost of Reducing Cloud Carbon Emissions,

    W. A. Hanafy, Q. Liang, N. Bashir, A. Souza, D. Irwin, and P. Shenoy, “Going Green for Less Green: Optimizing the Cost of Reducing Cloud Carbon Emissions,” in Proceedings of the 29th ACM International Conference on Architectural Support for Programming Languages and Operating ...

    2024

  28. [36]

    Ecovisor: A Virtual Energy System for Carbon-Efficient Applications,

    A. Souza, N. Bashir, J. Murillo, W. Hanafy, Q. Liang, D. Irwin, and P. Shenoy, “Ecovisor: A Virtual Energy System for Carbon-Efficient Applications,” in Proceedings of the 28th ACM International Conference on Architectural Support for Programming Languages and Operating System...

    2023

  29. [37]

    High-speed transfer optimization based on historical analysis and real-time tuning,

    E. Arslan and T. Kosar, “High-speed transfer optimization based on historical analysis and real-time tuning,” IEEE Transactions on Parallel and Distributed Systems , vol. 29, no. 6, pp. 1303–1316, 2018

    2018

  30. [38]

    Kosar, Data placement in widely distributed systems

    T. Kosar, Data placement in widely distributed systems . The University of Wisconsin-Madison, 2005

    2005

  31. [39]

    Disc: A system for distributed data intensive scientific computing

    G. Kola, T. Kosar, J. Frey, M. Livny, R. Brunner, and M. Remijan, “Disc: A system for distributed data intensive scientific computing.” in WORLDS, 2004

    2004

  32. [40]

    Greencourier: Carbon-aware scheduling for serverless functions,

    M. Chadha, T. Subramanian, E. Arima, M. Gerndt, M. Schulz, and O. Abboud, “Greencourier: Carbon-aware scheduling for serverless functions,” Proceedings of the 9th International Workshop on Serverless Computing, 2023. [Online]. Available: https://dl.acm.org/doi/10.1145/ 3631295.3631396

    2023

  33. [41]

    Caribou: Fine-grained geospatial shifting of serverless applications for sustainability

    V . Gsteiger, J. Sun, P. Javanrood, and M. Shahrad, “Caribou: Fine-grained geospatial shifting of serverless applications for sustainability.” [Online]. Available: https://api.semanticscholar.org/ CorpusID:273692349

  34. [42]

    Towards a Green Approach for Min- imizing Carbon Emissions in Fog-Cloud Architecture,

    M. Aldossary and H. A. Alharbi, “Towards a Green Approach for Min- imizing Carbon Emissions in Fog-Cloud Architecture,” IEEE Access , vol. 9, pp. 131 720–131 732, 2021

    2021

  35. [43]

    Carbonedge: Leveraging mesoscale spatial carbon-intensity variations for low carbon edge computing,

    L. Wu, W. A. Hanafy, A. Souza, K. Nguyen, J. Harkes, D. Irwin, M. Satyanarayanan, and P. Shenoy, “Carbonedge: Leveraging mesoscale spatial carbon-intensity variations for low carbon edge computing,”

  36. [44]

    Chv ´atal, Linear Programming

    V . Chv ´atal, Linear Programming . Macmillan, 9 1983. [Online]. Available: https://books.google.com/books/about/Linear Programming. html?hl=&id=DN20 tW BV0C

    1983

  37. [45]

    Energy-aware data transfer al- gorithms,

    I. Alan, E. Arslan, and T. Kosar, “Energy-aware data transfer al- gorithms,” in Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis , ser. SC ’15. New York, NY , USA: Association for Computing Machinery, 2015

    2015

  38. [46]

    Lessons learned from the chameleon testbed,

    K. Keahey, J. Anderson, Z. Zhen, P. Riteau, P. Ruth, D. Stanzione, M. Cevik, J. Colleran, H. S. Gunawi, C. Hammock, J. Mambretti, A. Barnes, F. Halbach, A. Rocha, and J. Stubbs, “Lessons learned from the chameleon testbed,” in Proceedings of the 2020 USENIX Annual Technical Co...

    2020

  39. [47]

    linprog; SciPy v1.15.2 Manual — docs.scipy.org,

    “linprog; SciPy v1.15.2 Manual — docs.scipy.org,” https://docs. scipy.org/doc/scipy/reference/generated/scipy.optimize.linprog.html, [Accessed 28-03-2025]

    2025

  40. [48]

    Data-driven algorithm selection for carbon-aware scheduling,

    R. Bostandoost, W. A. Hanafy, A. Lechowicz, N. Bashir, P. Shenoy, and M. Hajiesmaili, “Data-driven algorithm selection for carbon-aware scheduling,” in Proceedings of the 3rd Workshop on Sustainable Com- puter Systems. HotCarbon , vol. 24, 2024

    2024

  41. [49]

    iPerf - The TCP, UDP and SCTP network bandwidth measurement tool — iperf.fr,

    V . Gueant, “iPerf - The TCP, UDP and SCTP network bandwidth measurement tool — iperf.fr,” https://iperf.fr/, [Accessed 28-03-2025]. 9

    2025

  42. [2022]

    Available: https://doi.org/10.1007/978-3-031-12597-3 14

    [Online]. Available: https://doi.org/10.1007/978-3-031-12597-3 14

    doi

  43. [2025]

    Available: https://arxiv.org/abs/2502.14076

    [Online]. Available: https://arxiv.org/abs/2502.14076

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.