DRCD: A Regional-Contention-Driven Arbitration Policy for CPU-GPU Heterogeneous Systems

DRCD: A Regional-Contention-Driven Arbitration Policy for CPU-GPU Heterogeneous Systems | 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 DRCD: A Regional-Contention-Driven Arbitration Policy for CPU-GPU Heterogeneous Systems Juan Fang, Haoyu Cheng, Yuening Wang, Ran Zhai This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5650442/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 24 You are reading this latest preprint version Abstract In CPU-GPU heterogeneous systems, there exists intense resource contention between CPUs and GPUs. Traditional resource arbitration policies fail to account for the heterogeneity of cores, leading to inefficient network resource utilization for the CPU, which negatively impacts its performance. In heterogeneous networks, the degree of resource contention varies across different regions. This paper first uses reinforcement learning to analyze the message feature weights relied upon for resource arbitration in different network regions. To achieve more efficient resource allocation, a regional-contention-driven arbitration policy is proposed. Simulation results show that, compared to traditional arbitration policy, the overall network latency is reduced by 7.99%, and CPU performance is improved by 11.42%. Furthermore, a dynamic regional-contention-driven arbitration policy is proposed, which further reduces the overall network latency by 10.47% and increases CPU performance by 16.79% compared to traditional arbitration policy. Heterogenous Architectures Network-on-chip(NoC) Arbitration Policy Machine Learning Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 06 Jan, 2025 Reviews received at journal 05 Jan, 2025 Reviews received at journal 03 Jan, 2025 Reviews received at journal 03 Jan, 2025 Reviewers agreed at journal 03 Jan, 2025 Reviewers agreed at journal 30 Dec, 2024 Reviewers agreed at journal 30 Dec, 2024 Reviews received at journal 29 Dec, 2024 Reviewers agreed at journal 27 Dec, 2024 Reviewers agreed at journal 27 Dec, 2024 Reviewers agreed at journal 26 Dec, 2024 Reviewers agreed at journal 26 Dec, 2024 Reviews received at journal 26 Dec, 2024 Reviewers agreed at journal 26 Dec, 2024 Reviewers agreed at journal 25 Dec, 2024 Reviewers agreed at journal 25 Dec, 2024 Reviewers agreed at journal 24 Dec, 2024 Reviewers agreed at journal 24 Dec, 2024 Reviewers agreed at journal 24 Dec, 2024 Reviewers agreed at journal 24 Dec, 2024 Reviewers invited by journal 24 Dec, 2024 Editor assigned by journal 18 Dec, 2024 Submission checks completed at journal 18 Dec, 2024 First submitted to journal 15 Dec, 2024 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-5650442","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":391801088,"identity":"f8b8e697-e40a-4955-8b53-dfada56d2bb1","order_by":0,"name":"Juan Fang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA6klEQVRIiWNgGAWjYBAC9gYgIQHE/BIMCSABxgZCWngOQLVIzgBqOUC0FhAwuAEkiNPCfvbwC8s2uzzj2w3PpD8w2MhuOMD87AFeLTx5aRaSbcnFZncOpEkcYEgz3nCAzdwAnxZ7hhwzA8k25sRtNxJAWg4nbjjAwyaB1xb+NyAt9YmbZ4C1/CdCi0SO8QPJNqDhEmAtB4jR8saMQeLc8cQZdw4kW5wxSDaeeZjNjIDDcow/S5RVJ/bP7km8UVFhJ9t3vPkZXi1AwCYNUcGTAIwdIM1MQD1IyccPYJr9AGG1o2AUjIJRMCIBAKS8S3y5UakUAAAAAElFTkSuQmCC","orcid":"","institution":"Beijing University of Technology","correspondingAuthor":true,"prefix":"","firstName":"Juan","middleName":"","lastName":"Fang","suffix":""},{"id":391801089,"identity":"abb38487-5e8d-4298-b665-eab5a47cb2c5","order_by":1,"name":"Haoyu Cheng","email":"","orcid":"","institution":"Beijing University of Technology","correspondingAuthor":false,"prefix":"","firstName":"Haoyu","middleName":"","lastName":"Cheng","suffix":""},{"id":391801090,"identity":"d3b7b115-14a3-4f04-8661-7810e97ce0ec","order_by":2,"name":"Yuening Wang","email":"","orcid":"","institution":"Beijing University of Technology","correspondingAuthor":false,"prefix":"","firstName":"Yuening","middleName":"","lastName":"Wang","suffix":""},{"id":391801092,"identity":"b517f4fb-9a74-4d9d-aae4-79a9cc2551d7","order_by":3,"name":"Ran Zhai","email":"","orcid":"","institution":"Beijing University of Technology","correspondingAuthor":false,"prefix":"","firstName":"Ran","middleName":"","lastName":"Zhai","suffix":""}],"badges":[],"createdAt":"2024-12-16 04:38:18","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5650442/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5650442/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":72023211,"identity":"43e79e9b-23b8-4b12-9065-1a956fd21711","added_by":"auto","created_at":"2024-12-20 17:52:34","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":951601,"visible":true,"origin":"","legend":"","description":"","filename":"DRCDARegionalContentionDrivenArbitrationPolicyforCPUGPUHeterogeneousSystems.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5650442/v1_covered_f0cffcb0-4ef0-4194-ab0b-b10bd8489ace.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"DRCD: A Regional-Contention-Driven Arbitration Policy for CPU-GPU Heterogeneous Systems","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":"the-journal-of-supercomputing","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"Learn more about [The Journal of Supercomputing](https://www.springer.com/journal/11227)","snPcode":"11227","submissionUrl":"https://submission.nature.com/new-submission/11227/3","title":"The Journal of Supercomputing","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Heterogenous Architectures, Network-on-chip(NoC), Arbitration Policy, Machine Learning","lastPublishedDoi":"10.21203/rs.3.rs-5650442/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5650442/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eIn CPU-GPU heterogeneous systems, there exists intense resource contention between CPUs and GPUs. Traditional resource arbitration policies fail to account for the heterogeneity of cores, leading to inefficient network resource utilization for the CPU, which negatively impacts its performance. In heterogeneous networks, the degree of resource contention varies across different regions. This paper first uses reinforcement learning to analyze the message feature weights relied upon for resource arbitration in different network regions. To achieve more efficient resource allocation, a regional-contention-driven arbitration policy is proposed. Simulation results show that, compared to traditional arbitration policy, the overall network latency is reduced by 7.99%, and CPU performance is improved by 11.42%. Furthermore, a dynamic regional-contention-driven arbitration policy is proposed, which further reduces the overall network latency by 10.47% and increases CPU performance by 16.79% compared to traditional arbitration policy.\u003c/p\u003e","manuscriptTitle":"DRCD: A Regional-Contention-Driven Arbitration Policy for CPU-GPU Heterogeneous Systems","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-12-20 17:20:28","doi":"10.21203/rs.3.rs-5650442/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-01-06T18:26:09+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-01-06T01:48:02+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-01-03T09:38:57+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-01-03T08:51:25+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"54563167678323214498877433249020228681","date":"2025-01-03T06:42:27+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"187246528824553735748799925424083162585","date":"2024-12-30T08:50:17+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"185165460815552538889403627943176317565","date":"2024-12-30T07:16:00+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-12-30T02:51:32+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"84097181472780118939847201343441547082","date":"2024-12-27T07:41:10+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"290416051590737586707736846667713856803","date":"2024-12-27T05:33:28+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"297531575074666716514045362149085497741","date":"2024-12-27T04:30:03+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"198832679845346307455229992206847203721","date":"2024-12-27T03:14:40+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-12-26T08:43:27+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"66896494861699219696626343483223755430","date":"2024-12-26T05:56:47+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"9929565589607075304860300796410895079","date":"2024-12-25T11:53:44+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"210644595400794978710010754153186700055","date":"2024-12-25T08:22:33+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"207241167276740899827680890967963048769","date":"2024-12-25T02:05:06+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"228199459049433227081278065621327172497","date":"2024-12-25T01:38:35+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"43716170036780540177150372418857541117","date":"2024-12-25T00:15:18+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"203884232329056886311936175356099624185","date":"2024-12-24T23:26:55+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2024-12-24T21:53:49+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-12-18T09:16:34+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-12-18T09:13:35+00:00","index":"","fulltext":""},{"type":"submitted","content":"The Journal of Supercomputing","date":"2024-12-16T04:32:26+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"the-journal-of-supercomputing","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"Learn more about [The Journal of Supercomputing](https://www.springer.com/journal/11227)","snPcode":"11227","submissionUrl":"https://submission.nature.com/new-submission/11227/3","title":"The Journal of Supercomputing","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"f517589c-71bc-45ed-a791-45a02ac9c282","owner":[],"postedDate":"December 20th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2025-01-27T18:38:22+00:00","versionOfRecord":[],"versionCreatedAt":"2024-12-20 17:20:28","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-5650442","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5650442","identity":"rs-5650442","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}