{ "schema_version": "0.6.0", "paper_id": "assessing-predictive-models-energy-consumption-software", "page_url": "https://threadslab.org/research-publications/papers/assessing-predictive-models-energy-consumption-software/", "title": "Assessing Predictive Models for Energy Consumption Across Varied Software Environments", "title_variants": [], "authors": [ "Tong Zhang", "Sarwat Islam Dipanzan", "Leila Tahmooresnejad", "Naser Ezzati-Jivan" ], "author_details": [ { "name": "Tong Zhang", "orcid": null, "profile_url": "https://dblp.org/pid/07/4227.html" }, { "name": "Sarwat Islam Dipanzan", "orcid": null, "profile_url": "https://dblp.org/pid/397/8548.html" }, { "name": "Leila Tahmooresnejad", "orcid": null, "profile_url": "https://dblp.org/pid/156/2580.html" }, { "name": "Naser Ezzati-Jivan", "orcid": "https://orcid.org/0000-0003-1435-6297", "profile_url": "https://threadslab.org/" } ], "publication": { "year": 2024, "venue": "2024 IEEE International Conference on Big Data (BigData)", "type": "conference paper", "publication_date": "2024-12-15", "online_date": null, "print_date": "2024-12-15", "volume": null, "issue": null, "pages": "5233-5242", "article_number": null, "publisher": "IEEE", "issn": [], "isbn": [], "crossref_type": "proceedings-article" }, "publication_type": "conference paper", "status": "published_metadata_record", "canonical_source_url": "https://doi.org/10.1109/BigData62323.2024.10825500", "source_record_id": "assessing-predictive-models-for-energy-consumption-across-varied-software-environments-1da35b5bb9", "identifiers": { "doi": "10.1109/BigData62323.2024.10825500" }, "abstract": null, "abstract_source": "OpenAlex abstract metadata and public ResearchGate record reviewed; abstract not reproduced because reuse permission for the publisher version was not established.", "abstract_available": false, "scholar_eligibility": { "eligible": false, "basis": "not-eligible", "note": "The page is a discovery record; it does not claim Google Scholar article-host eligibility." }, "description": "The paper evaluates whether software-energy predictors transfer across applications when they use hardware-performance and operating-system event representations rather than application-specific measurements alone.", "evidence_level": "full-text-reviewed", "evidence": { "source_basis": "full-text-reviewed", "coverage": "material paper sections", "summary_origin": "AI-assisted catalog editorial summary", "review_status": "catalog-reviewed; paper-author approval pending", "verified_on": "2026-08-26", "sources": [ { "note": "Local PDF reviewed, including hardware, workload matrix, model comparison, and cross-application evaluation" }, { "note": "IEEE DOI record: https://doi.org/10.1109/BigData62323.2024.10825500" }, { "note": "IEEE Xplore record: https://ieeexplore.ieee.org/document/10825500/" } ] }, "summary": { "core_contribution": "The paper evaluates whether software-energy predictors transfer across applications when they use hardware-performance and operating-system event representations rather than application-specific measurements alone.", "problem": "Energy models are sensitive to the selected counters, operating-system behavior, hardware, and workload. A model that performs well within one application can therefore be misleading when used on a different software environment or on unseen data.", "method": "The study measures energy with Intel RAPL and collects performance counters with perf on a Ryzen 7 5825U system running Ubuntu 22.04.2 at the stated 1.6 GHz configuration. Stress-ng supplies controlled CPU workloads with 32 tasks and 6,528 combinations; Firefox versions 91, 96, 101, 105, 106, and 111 and MongoDB 7.0.0–7.0.4 provide application cases. The evaluated regressors include linear regression, LASSO, Ridge, SGD, MLP, and random forest, with repeated measurements, MAPE evaluation, and comparisons between event-based and alternative feature representations.", "findings": "The results show no universal best model: within-application performance is often strongest for hybrid feature sets, while event-based representations are more promising for transfer to unseen software conditions. The reported application-level MAPE values include 7.2%, 4.51%, and 6.58% for Firefox cases and 2.91% for the MongoDB case, but the error can exceed 100% for some model/feature combinations. The central empirical result is therefore the dependence of prediction quality on the application and feature representation, not a single universally transferable predictor.", "limitations": "The measurements use one laptop-class CPU, one operating-system configuration, Intel RAPL, selected software versions, and controlled workloads. Cross-application splits, counter availability, frequency behavior, and model hyperparameters constrain external validity; software energy outside the tested inference/workload conditions is not established.", "future_work": "Replicate the comparison on heterogeneous CPUs, GPUs, operating systems, frequency policies, and larger application suites; report confidence intervals and leakage-safe cross-application splits; and evaluate online calibration under workload drift and concurrent activity." }, "tags": [ "energy-efficiency", "performance-modeling", "predictive-monitoring", "machine-learning" ], "keywords": [ "software energy consumption", "predictive models", "energy efficiency", "software environments", "IEEE Big Data 2024" ], "versions": [ { "id": "published-version", "label": "Published version", "relation": "version-of-record", "title": "Assessing Predictive Models for Energy Consumption Across Varied Software Environments", "url": "https://doi.org/10.1109/BigData62323.2024.10825500", "pdf_url": null, "status": "published", "canonical_for_citation": true }, { "id": "ieee-record", "label": "IEEE record", "relation": "source-record", "title": "Assessing Predictive Models for Energy Consumption Across Varied Software Environments", "url": "https://ieeexplore.ieee.org/document/10825500/", "pdf_url": null, "status": "public_source_record", "canonical_for_citation": false } ], "access": { "status": "published_metadata_record", "note": "Crossref, IEEE, DBLP, and the proceedings table of contents verify the identity. The IEEE PDF was not available through normal access, so the record remains metadata-only.", "license": null }, "resources": { "code": "https://github.com/antoniotongzhang/IEEE%20BigData%202024", "data": null, "slides": null, "demo": null }, "citation_guidance": { "when_to_cite": "Cite this paper when evaluating whether software-energy models transfer across applications or feature representations.", "points": [ "Perf/RAPL-calibrated regression and neural energy predictors.", "Event-based versus alternative features for unseen-software generalization.", "The Firefox/MongoDB and Stress-ng evaluation design.", "The finding that no single model is universally best across software environments." ], "canonical_version_id": "published-version" }, "provenance": { "metadata_verified_on": "2026-08-26", "metadata_source": [ "Local PDF reviewed, including hardware, workload matrix, model comparison, and cross-application evaluation", "IEEE DOI record: https://doi.org/10.1109/BigData62323.2024.10825500", "IEEE Xplore record: https://ieeexplore.ieee.org/document/10825500/" ], "summary_written_by": "AI-assisted", "summary_verified_by": "full-text-grounded catalog review; author approval pending", "linked_preprint_record": null, "author_order_note": null }, "batch": { "phase": 2, "batch_label": "expanded forty-paper release", "status": "included_in_expanded_catalog", "selected_at": "2026-08-28" } }