Dyna-P: Placement-aware Dynamic Partitioning for Lightweight Applications with Modern GPUs

Dyna-P: Placement-aware Dynamic Partitioning for Lightweight Applications with Modern GPUs | 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 Dyna-P: Placement-aware Dynamic Partitioning for Lightweight Applications with Modern GPUs Theodora Adufu, Yoonhee Kim This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5774593/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 12 You are reading this latest preprint version Abstract Efficient GPU resource sharing is critical in dynamic cloud-based environments, particularly for lightweight HPC applications and Small Language Models, which demand partial GPU resources for execution. However, traditional scheduling frameworks fail to address intra-GPU and inter-node resource fragmentation and dynamic placement challenges arising from the heterogeneity in each application's resource demand and job completion times. This leads to resource under-utilization and scheduling delays in GPU clusters. This paper introduces Dyna-P, a novel scheduling framework designed to dynamically adjust GPU partitions to minimize resource fragmentation while improving system throughput and Makespan. Dyna-P proposes a Reconfiguration Last Policy which recognizes that workloads consisting of lightweight applications can benefit more from uninterrupted execution. Experimental results demonstrate that Dyna-P improves average throughput by up to 14.7% and reduces Makespan by 39% compared to state-of-the-art methods. These findings underscore Dyna-P’s potential to improve resource allocation rates in multi-tenant GPU environments Dynamic Partitioning Spatial Sharing GPU utilization Placement Fragmentation Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 26 Feb, 2025 Reviews received at journal 23 Feb, 2025 Reviews received at journal 21 Feb, 2025 Reviewers agreed at journal 17 Feb, 2025 Reviewers agreed at journal 14 Feb, 2025 Reviewers agreed at journal 13 Feb, 2025 Reviewers agreed at journal 11 Feb, 2025 Reviewers agreed at journal 14 Jan, 2025 Reviewers invited by journal 13 Jan, 2025 Editor assigned by journal 12 Jan, 2025 Submission checks completed at journal 09 Jan, 2025 First submitted to journal 06 Jan, 2025 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-5774593","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":400971883,"identity":"0eb3344f-5270-4b2b-8c11-1ce8d86aa1fb","order_by":0,"name":"Theodora Adufu","email":"","orcid":"","institution":"Sookmyung Women's University","correspondingAuthor":false,"prefix":"","firstName":"Theodora","middleName":"","lastName":"Adufu","suffix":""},{"id":400971884,"identity":"8a2d5c2e-56b0-4633-aa2d-78eabe3655cb","order_by":1,"name":"Yoonhee Kim","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA4UlEQVRIiWNgGAWjYFAC5gYGhgogzYMkREALI1DLGZK1MLaRosXgRmLj58J5dXL8PKfTpHl31DLwtx9gNq7Ar6VZeua2w8aSvb3bpHnPHGeQOJPAnHgGv5YGad5tBxI3nOcFamk7xsBwg4H5YAMBW37zzqlDaJEnQkubNG8Dc+KGsyCHtdUARRiYE/FpkTzzsM2a5xjQLz1nN1vObTvAY3gmsdkQnxa+48mHb/PUgEIsd+ONt211cnLHDx+WxKdF4QCCzSLBwHCYBxK5eIA8kjTzBwaGOryqR8EoGAWjYGQCAFUuTqqB48UVAAAAAElFTkSuQmCC","orcid":"","institution":"Sookmyung Women's University","correspondingAuthor":true,"prefix":"","firstName":"Yoonhee","middleName":"","lastName":"Kim","suffix":""}],"badges":[],"createdAt":"2025-01-06 14:23:19","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5774593/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5774593/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":73615635,"identity":"15b6ca58-092c-4658-b16a-882258a0a26d","added_by":"auto","created_at":"2025-01-13 02:44:30","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1912950,"visible":true,"origin":"","legend":"","description":"","filename":"DynaPrevised.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5774593/v1_covered_d8e83300-0553-4f3e-8a32-8e7fd50a4ec2.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Dyna-P: Placement-aware Dynamic Partitioning for Lightweight Applications with Modern GPUs","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"cluster-computing","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"Learn more about [Cluster Computing](https://www.springer.com/journal/10586)","snPcode":"10586","submissionUrl":"https://submission.nature.com/new-submission/10586/3","title":"Cluster Computing","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Dynamic Partitioning, Spatial Sharing, GPU utilization, Placement, Fragmentation","lastPublishedDoi":"10.21203/rs.3.rs-5774593/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5774593/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Efficient GPU resource sharing is critical in dynamic cloud-based environments, particularly for lightweight HPC applications and Small Language Models, which demand partial GPU resources for execution. However, traditional scheduling frameworks fail to address intra-GPU and inter-node resource fragmentation and dynamic placement challenges arising from the heterogeneity in each application's resource demand and job completion times. This leads to resource under-utilization and scheduling delays in GPU clusters. This paper introduces Dyna-P, a novel scheduling framework designed to dynamically adjust GPU partitions to minimize resource fragmentation while improving system throughput and Makespan. Dyna-P proposes a Reconfiguration Last Policy which recognizes that workloads consisting of lightweight applications can benefit more from uninterrupted execution. Experimental results demonstrate that Dyna-P improves average throughput by up to 14.7% and reduces Makespan by 39% compared to state-of-the-art methods. These findings underscore Dyna-P’s potential to improve resource allocation rates in multi-tenant GPU environments","manuscriptTitle":"Dyna-P: Placement-aware Dynamic Partitioning for Lightweight Applications with Modern GPUs","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-01-13 02:20:24","doi":"10.21203/rs.3.rs-5774593/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-02-26T17:13:53+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-02-23T15:26:02+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-02-21T13:39:17+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"98911542571870886637099565855842999441","date":"2025-02-17T12:27:33+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"262597933470572414147910226551924144783","date":"2025-02-14T21:30:36+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"161648915085915500824200751781024308317","date":"2025-02-14T01:33:52+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"284805356923684863163617514865176899498","date":"2025-02-11T20:43:25+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"20253298305506146981428113037586663916","date":"2025-01-14T05:23:48+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-01-13T05:28:41+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-01-12T06:05:39+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-01-09T13:38:40+00:00","index":"","fulltext":""},{"type":"submitted","content":"Cluster Computing","date":"2025-01-06T14:13:37+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"cluster-computing","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"Learn more about [Cluster Computing](https://www.springer.com/journal/10586)","snPcode":"10586","submissionUrl":"https://submission.nature.com/new-submission/10586/3","title":"Cluster Computing","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"96c8611e-758f-4c86-8312-15ea8a0a71a3","owner":[],"postedDate":"January 13th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2025-04-05T22:08:11+00:00","versionOfRecord":[],"versionCreatedAt":"2025-01-13 02:20:24","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-5774593","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5774593","identity":"rs-5774593","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}