Priority-Aware Resource Reallocation in Edge Computing Using Reinforcement Learning

Priority-Aware Resource Reallocation in Edge Computing Using Reinforcement Learning | 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 Priority-Aware Resource Reallocation in Edge Computing Using Reinforcement Learning Sudabeh Mohammadi, Behzad Akbari This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6545669/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 31 Oct, 2025 Read the published version in Journal of Network and Systems Management → Version 1 posted 11 You are reading this latest preprint version Abstract The rapid adoption of IoT applications has led to the continuous generation of vast amounts of data, demanding efficient processing, storage, and real-time response delivery. Edge computing has emerged as a critical technology that addresses the growing need for low-latency and high-bandwidth applications in the Internet of Things (IoT) ecosystem. By processing data closer to its source, edge computing reduces dependence on centralized data centers, significantly improving response times.However, this shift introduces significant challenges in resource management, particularly in allocating limited and heterogeneous computational, storage, and network resources across distributed edge nodes. Existing solutions often fail to adapt to real-time priority shifts or enforce strict Quality of Service (QoS) guarantees for critical tasks (e.g., healthcare, real-time gaming). To address these challenges, this paper proposes two novel reinforcement learning (RL)-based resource reallocation algorithms that dynamically optimize edge resource allocation by: 1- Priority-aware task classification: Categorizing tasks into five demand-based levels (e.g., bandwidth-intensive, time-sensitive) and three priority classes (critical, important, general), enabling context-aware decision-making. 2- Dynamic preemption: Reallocating resources from low-priority tasks to high-priority ones while minimizing disruptions to ongoing processes. 3- MDP-based optimization: Formulating the NP-hard resource allocation problem as a Markov Decision Process (MDP) and solving it via Q-learning, prioritizing time-sensitive tasks.Simulations demonstrate that our approach reduces task rejection rates by up to 30% in critical tasks and 2% in important tasks compared to baseline methods, while ensuring 80% acceptance of critical tasks. Results demonstrate effective efficiency-QoS tradeoffs in a dynamic edge environment. Acceptance rate Edge computing Reallocation algorithm Reinforcement learning Resource allocation Response time Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 31 Oct, 2025 Read the published version in Journal of Network and Systems Management → Version 1 posted Editorial decision: Revision requested 24 Jul, 2025 Reviews received at journal 10 Jul, 2025 Reviews received at journal 06 Jul, 2025 Reviews received at journal 29 Jun, 2025 Reviewers agreed at journal 20 Jun, 2025 Reviewers agreed at journal 12 Jun, 2025 Reviewers agreed at journal 10 Jun, 2025 Reviewers invited by journal 10 Jun, 2025 Editor assigned by journal 04 Jun, 2025 Submission checks completed at journal 29 Apr, 2025 First submitted to journal 28 Apr, 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-6545669","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":469454506,"identity":"46249dfb-c97d-46e4-98dc-3f4a62fe92b0","order_by":0,"name":"Sudabeh Mohammadi","email":"","orcid":"","institution":"Tarbiat Modares University","correspondingAuthor":false,"prefix":"","firstName":"Sudabeh","middleName":"","lastName":"Mohammadi","suffix":""},{"id":469454508,"identity":"5670076f-93fe-4214-9319-7287bf7cfe46","order_by":1,"name":"Behzad Akbari","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA7klEQVRIiWNgGAWjYBACAwYGZmYgzcMPJCTgwgnEaJFsIFULg8EBZC34gDkD82Pjgpp7MsbXDh+88XOPjT0D++EHDA/34NZi2cBmnDzjWDGP2e20ZMueZ2mJDTxpBgwJz/A47ACD8WEetgSglhwzCZ4Dh4G+yAH65QA+LeyfD/P8S+Axnp1jJvnnwH97Bv43hLTwGCfztiXwGEjnmEnzHDjA2CBBwBbLZp5i45l9CTwSQL9YyxxITmyTeGZwAJ8Wc/b2zdIF3xLs+WcnH7z55oCdPT9/8sOHP/BoYWBGF2ADYnwaRsEoGAWjYBQQAQCTB0pgBDk7WwAAAABJRU5ErkJggg==","orcid":"","institution":"Tarbiat Modares University","correspondingAuthor":true,"prefix":"","firstName":"Behzad","middleName":"","lastName":"Akbari","suffix":""}],"badges":[],"createdAt":"2025-04-28 08:38:10","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6545669/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6545669/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s10922-025-09992-3","type":"published","date":"2025-10-31T15:57:52+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":95040454,"identity":"1ee9c810-df7c-486a-b8b7-192af188b531","added_by":"auto","created_at":"2025-11-03 16:08:58","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":606179,"visible":true,"origin":"","legend":"","description":"","filename":"NetworkandSystemsManagement.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6545669/v1_covered_e662d02c-17e0-4807-9283-e966ba75021f.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Priority-Aware Resource Reallocation in Edge Computing Using Reinforcement Learning","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":"journal-of-network-and-systems-management","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"jons","sideBox":"Learn more about [Journal of Network and Systems Management](http://link.springer.com/journal/10922)","snPcode":"10922","submissionUrl":"https://submission.nature.com/new-submission/10922/3","title":"Journal of Network and Systems Management","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Acceptance rate, Edge computing, Reallocation algorithm, Reinforcement learning, Resource allocation, Response time","lastPublishedDoi":"10.21203/rs.3.rs-6545669/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6545669/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe rapid adoption of IoT applications has led to the continuous generation of vast amounts of data, demanding efficient processing, storage, and real-time response delivery. Edge computing has emerged as a critical technology that addresses the growing need for low-latency and high-bandwidth applications in the Internet of Things (IoT) ecosystem. By processing data closer to its source, edge computing reduces dependence on centralized data centers, significantly improving response times.However, this shift introduces significant challenges in resource management, particularly in allocating limited and heterogeneous computational, storage, and network resources across distributed edge nodes. Existing solutions often fail to adapt to real-time priority shifts or enforce strict Quality of Service (QoS) guarantees for critical tasks (e.g., healthcare, real-time gaming). To address these challenges, this paper proposes two novel reinforcement learning (RL)-based resource reallocation algorithms that dynamically optimize edge resource allocation by: 1- Priority-aware task classification: Categorizing tasks into five demand-based levels (e.g., bandwidth-intensive, time-sensitive) and three priority classes (critical, important, general), enabling context-aware decision-making. 2- Dynamic preemption: Reallocating resources from low-priority tasks to high-priority ones while minimizing disruptions to ongoing processes. 3- MDP-based optimization: Formulating the NP-hard resource allocation problem as a Markov Decision Process (MDP) and solving it via Q-learning, prioritizing time-sensitive tasks.Simulations demonstrate that our approach reduces task rejection rates by up to 30% in critical tasks and 2% in important tasks compared to baseline methods, while ensuring 80% acceptance of critical tasks. Results demonstrate effective efficiency-QoS tradeoffs in a dynamic edge environment.\u003c/p\u003e","manuscriptTitle":"Priority-Aware Resource Reallocation in Edge Computing Using Reinforcement Learning","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-06-12 05:31:03","doi":"10.21203/rs.3.rs-6545669/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-07-24T12:46:53+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-07-10T22:34:22+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-07-06T09:47:36+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-06-29T16:07:56+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"37686972769435965200207239968776039766","date":"2025-06-21T02:51:57+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"328739969206492340901153395963845975799","date":"2025-06-12T17:13:30+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"287018591842117842449366758088104678675","date":"2025-06-10T16:10:04+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-06-10T16:05:22+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-06-04T07:33:22+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-04-29T04:47:48+00:00","index":"","fulltext":""},{"type":"submitted","content":"Journal of Network and Systems Management","date":"2025-04-28T08:23:30+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"journal-of-network-and-systems-management","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"jons","sideBox":"Learn more about [Journal of Network and Systems Management](http://link.springer.com/journal/10922)","snPcode":"10922","submissionUrl":"https://submission.nature.com/new-submission/10922/3","title":"Journal of Network and Systems Management","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"88f10a0b-696d-4806-8529-d41abb0c6d4c","owner":[],"postedDate":"June 12th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-11-03T16:04:53+00:00","versionOfRecord":{"articleIdentity":"rs-6545669","link":"https://doi.org/10.1007/s10922-025-09992-3","journal":{"identity":"journal-of-network-and-systems-management","isVorOnly":false,"title":"Journal of Network and Systems Management"},"publishedOn":"2025-10-31 15:57:52","publishedOnDateReadable":"October 31st, 2025"},"versionCreatedAt":"2025-06-12 05:31:03","video":"","vorDoi":"10.1007/s10922-025-09992-3","vorDoiUrl":"https://doi.org/10.1007/s10922-025-09992-3","workflowStages":[]},"version":"v1","identity":"rs-6545669","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6545669","identity":"rs-6545669","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}