DS-LSM-tree: SSTable-Decoupled LSM-tree

DS-LSM-tree: SSTable-Decoupled LSM-tree | 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 DS-LSM-tree: SSTable-Decoupled LSM-tree Tao Cai, Yikang Deng, Dejiao Niu, Zeyu Huang, Qiujing Huang, Maotao Jiang This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8587363/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 13 You are reading this latest preprint version Abstract The Log-Structured Merge Tree (LSM-tree) is widely adopted in key–value (KV) storage systems due to its ability to transform random writes into sequential writes. However, LSM-tree still suffers from issues such as write amplification and periodic performance degradation, which become more severe on emerging storage devices with strict sequential-write constraints, such as Zoned Namespace Solid State Drives (ZNS SSDs). Meanwhile, the introduction of Non-Volatile Memory (NVM) provides new opportunities for optimizing the internal structure of LSM-tree.We observe that traditional SSTables tightly couple metadata with data blocks, causing frequent metadata updates to trigger unnecessary rewrites of large data regions, thereby exacerbating write amplification. To address this problem, we propose DS-LSM-tree, an LSM-tree design based on SSTable decoupling. In DS-LSM-tree, large and infrequently accessed data blocks are sequentially stored on ZNS SSDs, while frequently accessed and fine-grained metadata are migrated to NVM, alleviating the mismatch between update granularity and storage device characteristics.Based on the decoupled SSTable design, we further develop a ZNS-aware SSTable metadata structure and propose an NVM-based logical merge strategy for SSTable data blocks, which reduces unnecessary physical data migrations. We implement a prototype system named SPMS based on DS-LSM-tree and evaluate it using YCSB benchmarks. Experimental results show that, compared with the traditional LSM-tree–based system ListDB, SPMS improves write throughput by up to 51.2%, increases read throughput by 5.1%, and reduces write amplification by 53.1%. LSM-tree ZNS SSD NVM Hybrid Storage Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 21 Apr, 2026 Reviews received at journal 19 Apr, 2026 Reviews received at journal 18 Apr, 2026 Reviews received at journal 18 Apr, 2026 Reviews received at journal 08 Apr, 2026 Reviewers agreed at journal 03 Apr, 2026 Reviewers agreed at journal 02 Apr, 2026 Reviewers agreed at journal 01 Apr, 2026 Reviewers agreed at journal 01 Apr, 2026 Reviewers invited by journal 31 Mar, 2026 Editor assigned by journal 31 Mar, 2026 Submission checks completed at journal 15 Jan, 2026 First submitted to journal 12 Jan, 2026 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-8587363","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":617139538,"identity":"4ac8f2d1-78f0-48cd-9333-f729be4cb173","order_by":0,"name":"Tao Cai","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAyUlEQVRIiWNgGAWjYFACxgYgYQOheIjU0ghUnEaSFrDiwxAmUVr42w+3P/hRcT6xuf8A44O3bQzy5oS0SJxJbGzsOXM7sXFGArPh3DYGw50NBLQYMCQ2NvC2gbQwsEnztjEkGBwgpIX/YWPj37ZziY39B9h/E6dFIrGxmbftANCuBDZmorRI3HjYOFvmTLJx44zEZsk55yQMNxDSwt+f/uDjmwo72Y39hw9+eFNmI0/QFjgwbABHpgSx6oFAngS1o2AUjIJRMMIAAAAtRDn05XcJAAAAAElFTkSuQmCC","orcid":"","institution":"JiangSu University","correspondingAuthor":true,"prefix":"","firstName":"Tao","middleName":"","lastName":"Cai","suffix":""},{"id":617139539,"identity":"978af5ca-e62f-420d-83ec-c9a52aa0bcc8","order_by":1,"name":"Yikang Deng","email":"","orcid":"","institution":"JiangSu University","correspondingAuthor":false,"prefix":"","firstName":"Yikang","middleName":"","lastName":"Deng","suffix":""},{"id":617139540,"identity":"de7d833d-b67d-4f7b-beea-15b7fa45e118","order_by":2,"name":"Dejiao Niu","email":"","orcid":"","institution":"JiangSu University","correspondingAuthor":false,"prefix":"","firstName":"Dejiao","middleName":"","lastName":"Niu","suffix":""},{"id":617139543,"identity":"ccf3cd86-efdb-4c58-aa27-bc26fa0a53d8","order_by":3,"name":"Zeyu Huang","email":"","orcid":"","institution":"JiangSu University","correspondingAuthor":false,"prefix":"","firstName":"Zeyu","middleName":"","lastName":"Huang","suffix":""},{"id":617139545,"identity":"d3e2fcb7-039e-4371-b9a2-000aa2670829","order_by":4,"name":"Qiujing Huang","email":"","orcid":"","institution":"JiangSu University","correspondingAuthor":false,"prefix":"","firstName":"Qiujing","middleName":"","lastName":"Huang","suffix":""},{"id":617139547,"identity":"c9e2882a-6bc4-47fa-bf5b-3f10dd67cdbc","order_by":5,"name":"Maotao Jiang","email":"","orcid":"","institution":"JiangSu University","correspondingAuthor":false,"prefix":"","firstName":"Maotao","middleName":"","lastName":"Jiang","suffix":""}],"badges":[],"createdAt":"2026-01-13 05:08:10","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8587363/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8587363/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":106403336,"identity":"45463cbb-d161-4496-be64-8f7839bcdd98","added_by":"auto","created_at":"2026-04-08 09:14:06","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":352566,"visible":true,"origin":"","legend":"","description":"","filename":"DSLSMtree.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8587363/v1_covered_0a9b0b42-6892-4731-8620-046e4243a5e4.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"DS-LSM-tree: SSTable-Decoupled LSM-tree","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"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":"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":"LSM-tree, ZNS SSD, NVM, Hybrid Storage","lastPublishedDoi":"10.21203/rs.3.rs-8587363/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8587363/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"The Log-Structured Merge Tree (LSM-tree) is widely adopted in key–value (KV) storage systems due to its ability to transform random writes into sequential writes. However, LSM-tree still suffers from issues such as write amplification and periodic performance degradation, which become more severe on emerging storage devices with strict sequential-write constraints, such as Zoned Namespace Solid State Drives (ZNS SSDs). Meanwhile, the introduction of Non-Volatile Memory (NVM) provides new opportunities for optimizing the internal structure of LSM-tree.We observe that traditional SSTables tightly couple metadata with data blocks, causing frequent metadata updates to trigger unnecessary rewrites of large data regions, thereby exacerbating write amplification. To address this problem, we propose DS-LSM-tree, an LSM-tree design based on SSTable decoupling. In DS-LSM-tree, large and infrequently accessed data blocks are sequentially stored on ZNS SSDs, while frequently accessed and fine-grained metadata are migrated to NVM, alleviating the mismatch between update granularity and storage device characteristics.Based on the decoupled SSTable design, we further develop a ZNS-aware SSTable metadata structure and propose an NVM-based logical merge strategy for SSTable data blocks, which reduces unnecessary physical data migrations. We implement a prototype system named SPMS based on DS-LSM-tree and evaluate it using YCSB benchmarks. Experimental results show that, compared with the traditional LSM-tree–based system ListDB, SPMS improves write throughput by up to 51.2\\%, increases read throughput by 5.1\\%, and reduces write amplification by 53.1\\%.","manuscriptTitle":"DS-LSM-tree: SSTable-Decoupled LSM-tree","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-06 07:03:55","doi":"10.21203/rs.3.rs-8587363/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-04-21T17:46:49+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-19T19:47:46+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-18T09:04:12+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-18T07:13:36+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-08T05:07:04+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"294107035334239683919238022200675364029","date":"2026-04-03T04:52:31+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"34580607207470617061856703741951635357","date":"2026-04-02T20:39:37+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"65498623889907261194477179195788565240","date":"2026-04-01T19:40:10+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"253788594470145866888040014012239436533","date":"2026-04-01T15:14:52+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-04-01T01:03:53+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-04-01T00:44:28+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-01-15T16:35:03+00:00","index":"","fulltext":""},{"type":"submitted","content":"The Journal of Supercomputing","date":"2026-01-13T04:57:16+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":"3cb4f732-f02a-4c19-ac49-5a734a0c7b71","owner":[],"postedDate":"April 6th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-05-14T03:23:37+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-06 07:03:55","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8587363","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8587363","identity":"rs-8587363","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}