Efficient Cluster Execution of Sparse Transformers: Joint Quantization and Carbon- Aware DVFS Scheduling

Efficient Cluster Execution of Sparse Transformers: Joint Quantization and Carbon- Aware DVFS Scheduling | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Efficient Cluster Execution of Sparse Transformers: Joint Quantization and Carbon- Aware DVFS Scheduling Kamal Saluja, Vikas Solanki, Sunil Gupta, Reema Goyal, Tanya Khaneja This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6689130/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract The rising computational demands of Transformer-based deep learning models pose significant challenges to energy efficiency, particularly in distributed cluster environments. This paper presents a novel framework for executing sparse Transformer models on compute clusters through the joint application of model quantization and carbon-aware dynamic voltage and frequency scaling (DVFS) scheduling. By leveraging structured sparsity and low-precision quantization techniques, we reduce memory footprint and computational overhead without compromising model accuracy. In parallel, a dynamic scheduling mechanism is proposed to optimize power usage based on carbon intensity forecasts and workload characteristics, ensuring environmentally sustainable inference. The framework is evaluated on benchmark Transformer architectures across simulated and real-world carbon-aware DVFS environments. Results demonstrate significant reductions in energy consumption—up to 45%—with negligible impact on accuracy, while enabling flexible adaptation to fluctuating energy availability. Our approach highlights a scalable, low-cost pathway for deploying deep learning models in resource-constrained and sustainability-sensitive computing clusters. This research bridges the gap between efficient AI inference and green computing, aligning with the emerging need for carbon-conscious, large-scale machine learning systems. Sparse Transformers Model Quantization Dynamic Voltage and Frequency Scaling (DVFS) Carbon-Aware Scheduling Energy Efficient Deep Learning Sustainable AI Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6689130","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":470353669,"identity":"5188ab68-ee38-4921-8276-cb131014f95f","order_by":0,"name":"Kamal Saluja","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABKElEQVRIie3QsUrDQBjA8e8INMtXu15Jq69QyWKhmle5EIhLM7l0KKVTsgRd27cIBDofHNTBgmsHh7hkshDoElDQuxZB5IK4Cd6f3AUu9+PCAZhMfzCqpkKOEwAGFXAEPH4hc/U0ESZHS77IQhL8DQELgcMXoq+bPPgVg9lZi7Ln/Wj61PNQlFUNo37GrbjQEAejnDIQ5zFlgTNel4jt2F2mELoZJ8lAQ04hyiThRBLmRHOB2AFXXoHwM7WoI51dXssf89Qpr8MDsffkDd4biUOjlTzF8iUJHaJIO3XVPTSS7mK3umADEcRYhMN0Lclmc2P1BoG7FHpCH6N8W01ml3f2ONjWU+HZ6XVOXiZX/dv7pNSRY4eLQfZtxWrc/5nNf9xiMplM/7MPi7hh2p/9G0wAAAAASUVORK5CYII=","orcid":"","institution":"Chitkara University Institute of Engineering and Technology, Chitkara University","correspondingAuthor":true,"prefix":"","firstName":"Kamal","middleName":"","lastName":"Saluja","suffix":""},{"id":470353670,"identity":"8a5210cf-b527-4fcb-bb14-252802f187c1","order_by":1,"name":"Vikas Solanki","email":"","orcid":"","institution":"Chitkara University Institute of Engineering and Technology, Chitkara University","correspondingAuthor":false,"prefix":"","firstName":"Vikas","middleName":"","lastName":"Solanki","suffix":""},{"id":470353671,"identity":"2359313f-810d-4ad1-8834-f6ee9e9c4b1a","order_by":2,"name":"Sunil Gupta","email":"","orcid":"","institution":"Chitkara University Institute of Engineering and Technology, Chitkara University","correspondingAuthor":false,"prefix":"","firstName":"Sunil","middleName":"","lastName":"Gupta","suffix":""},{"id":470353672,"identity":"f9c47b1d-842d-48ef-b615-53879f4b50d7","order_by":3,"name":"Reema Goyal","email":"","orcid":"","institution":"Chitkara University Institute of Engineering and Technology, Chitkara University","correspondingAuthor":false,"prefix":"","firstName":"Reema","middleName":"","lastName":"Goyal","suffix":""},{"id":470353673,"identity":"e64869bf-5ac0-4b12-9bb6-54e0b7b703ca","order_by":4,"name":"Tanya Khaneja","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Tanya","middleName":"","lastName":"Khaneja","suffix":""}],"badges":[],"createdAt":"2025-05-17 23:23:06","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6689130/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6689130/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":103587249,"identity":"3b77fc72-0d05-4845-adcf-ea83747fe694","added_by":"auto","created_at":"2026-02-27 11:27:31","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":656631,"visible":true,"origin":"","legend":"","description":"","filename":"cloudcomputingfinal.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6689130/v1_covered_b1fbac23-1e26-44bd-bffd-ba0ea6a47d27.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Efficient Cluster Execution of Sparse Transformers: Joint Quantization and Carbon- Aware DVFS Scheduling","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Sparse Transformers, Model Quantization, Dynamic Voltage and Frequency Scaling (DVFS), Carbon-Aware Scheduling, Energy Efficient Deep Learning, Sustainable AI","lastPublishedDoi":"10.21203/rs.3.rs-6689130/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6689130/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe rising computational demands of Transformer-based deep learning models pose significant challenges to energy efficiency, particularly in distributed cluster environments. This paper presents a novel framework for executing sparse Transformer models on compute clusters through the joint application of model quantization and carbon-aware dynamic voltage and frequency scaling (DVFS) scheduling. By leveraging structured sparsity and low-precision quantization techniques, we reduce memory footprint and computational overhead without compromising model accuracy. In parallel, a dynamic scheduling mechanism is proposed to optimize power usage based on carbon intensity forecasts and workload characteristics, ensuring environmentally sustainable inference. The framework is evaluated on benchmark Transformer architectures across simulated and real-world carbon-aware DVFS environments. Results demonstrate significant reductions in energy consumption\u0026mdash;up to 45%\u0026mdash;with negligible impact on accuracy, while enabling flexible adaptation to fluctuating energy availability. Our approach highlights a scalable, low-cost pathway for deploying deep learning models in resource-constrained and sustainability-sensitive computing clusters. This research bridges the gap between efficient AI inference and green computing, aligning with the emerging need for carbon-conscious, large-scale machine learning systems.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e","manuscriptTitle":"Efficient Cluster Execution of Sparse Transformers: Joint Quantization and Carbon- Aware DVFS Scheduling","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-06-13 14:42:54","doi":"10.21203/rs.3.rs-6689130/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"288351ca-bf13-4bcc-83bb-c5fbeff0e38d","owner":[],"postedDate":"June 13th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-02-27T11:26:34+00:00","versionOfRecord":[],"versionCreatedAt":"2025-06-13 14:42:54","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6689130","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6689130","identity":"rs-6689130","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}