Efficiently Enlarging RDMA-Attached Memory with SSD
RDMA-based in-memory storage systems offer high performance but are restricted by the capacity of physical memory. In this paper, we propose TeRM to extend RDMA-attached memory with SSD. TeRM achieves fast remote access on the SSD-extended memory by eliminating page faults of RDMA NIC and CPU from t...
Gespeichert in:
| Veröffentlicht in: | ACM transactions on storage 2025-05, Vol.21 (2), p.1-27 |
|---|---|
| Hauptverfasser: | , , , , , |
| Format: | Artikel |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | 27 |
|---|---|
| container_issue | 2 |
| container_start_page | 1 |
| container_title | ACM transactions on storage |
| container_volume | 21 |
| creator | Yang, Zhe Wang, Qing Liao, Xiaojian Lu, Youyou Huang, Keji Shu, Jiwu |
| description | RDMA-based in-memory storage systems offer high performance but are restricted by the capacity of physical memory. In this paper, we propose TeRM to extend RDMA-attached memory with SSD. TeRM achieves fast remote access on the SSD-extended memory by eliminating page faults of RDMA NIC and CPU from the critical path. We also introduce a set of techniques to reduce the consumption of CPU and network resources. Evaluation shows that TeRM performs close to the performance of the ideal upper bound where all pages are pinned in the physical memory. Compared with existing approaches TeRM significantly improves the performance of unmodified RDMA-based storage systems, including a file system and a key-value system. |
| doi_str_mv | 10.1145/3700772 |
| format | Article |
| fullrecord | <record><control><sourceid>acm_cross</sourceid><recordid>TN_cdi_crossref_primary_10_1145_3700772</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>3700772</sourcerecordid><originalsourceid>FETCH-LOGICAL-a842-e3ea605ee2c4282a99fd74a5f82503347980e8a68c57966bea3c99d38dbd053e3</originalsourceid><addsrcrecordid>eNo9j8FLwzAYxYMoOKd499Sbp-rXfEmTHMvWqbAhuN1Lln7ZKm0nSUH636ts7vQevB8PfozdZ_CUZUI-owJQil-wSSYlpggGL89dqWt2E-MnAOZcyAnD0vvGNdQP7ZiUfWvDrul3ycd8VaTFMFi3pzpZUXcIY_LdDPtkvZ7fsitv20h3p5yyzaLczF7T5fvL26xYplYLnhKSzUEScSe45tYYXythpddcAqJQRgNpm2snlcnzLVl0xtSo620NEgmn7PF468IhxkC--gpNZ8NYZVD9qVYn1V_y4Uha152h__EHwq1L2A</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Efficiently Enlarging RDMA-Attached Memory with SSD</title><source>ACM Digital Library Complete</source><creator>Yang, Zhe ; Wang, Qing ; Liao, Xiaojian ; Lu, Youyou ; Huang, Keji ; Shu, Jiwu</creator><creatorcontrib>Yang, Zhe ; Wang, Qing ; Liao, Xiaojian ; Lu, Youyou ; Huang, Keji ; Shu, Jiwu</creatorcontrib><description>RDMA-based in-memory storage systems offer high performance but are restricted by the capacity of physical memory. In this paper, we propose TeRM to extend RDMA-attached memory with SSD. TeRM achieves fast remote access on the SSD-extended memory by eliminating page faults of RDMA NIC and CPU from the critical path. We also introduce a set of techniques to reduce the consumption of CPU and network resources. Evaluation shows that TeRM performs close to the performance of the ideal upper bound where all pages are pinned in the physical memory. Compared with existing approaches TeRM significantly improves the performance of unmodified RDMA-based storage systems, including a file system and a key-value system.</description><identifier>ISSN: 1553-3077</identifier><identifier>EISSN: 1553-3093</identifier><identifier>DOI: 10.1145/3700772</identifier><language>eng</language><publisher>New York, NY: ACM</publisher><subject>Distributed memory ; Information storage systems ; Information systems ; Software and its engineering</subject><ispartof>ACM transactions on storage, 2025-05, Vol.21 (2), p.1-27</ispartof><rights>Copyright held by the owner/author(s).</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><cites>FETCH-LOGICAL-a842-e3ea605ee2c4282a99fd74a5f82503347980e8a68c57966bea3c99d38dbd053e3</cites><orcidid>0000-0002-6214-5390 ; 0000-0002-2031-5937 ; 0000-0002-7362-2789 ; 0000-0002-5526-7154 ; 0009-0009-5496-5522 ; 0000-0002-7924-9268</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>314,776,780,27901,27902</link.rule.ids></links><search><creatorcontrib>Yang, Zhe</creatorcontrib><creatorcontrib>Wang, Qing</creatorcontrib><creatorcontrib>Liao, Xiaojian</creatorcontrib><creatorcontrib>Lu, Youyou</creatorcontrib><creatorcontrib>Huang, Keji</creatorcontrib><creatorcontrib>Shu, Jiwu</creatorcontrib><title>Efficiently Enlarging RDMA-Attached Memory with SSD</title><title>ACM transactions on storage</title><addtitle>ACM TOS</addtitle><description>RDMA-based in-memory storage systems offer high performance but are restricted by the capacity of physical memory. In this paper, we propose TeRM to extend RDMA-attached memory with SSD. TeRM achieves fast remote access on the SSD-extended memory by eliminating page faults of RDMA NIC and CPU from the critical path. We also introduce a set of techniques to reduce the consumption of CPU and network resources. Evaluation shows that TeRM performs close to the performance of the ideal upper bound where all pages are pinned in the physical memory. Compared with existing approaches TeRM significantly improves the performance of unmodified RDMA-based storage systems, including a file system and a key-value system.</description><subject>Distributed memory</subject><subject>Information storage systems</subject><subject>Information systems</subject><subject>Software and its engineering</subject><issn>1553-3077</issn><issn>1553-3093</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2025</creationdate><recordtype>article</recordtype><recordid>eNo9j8FLwzAYxYMoOKd499Sbp-rXfEmTHMvWqbAhuN1Lln7ZKm0nSUH636ts7vQevB8PfozdZ_CUZUI-owJQil-wSSYlpggGL89dqWt2E-MnAOZcyAnD0vvGNdQP7ZiUfWvDrul3ycd8VaTFMFi3pzpZUXcIY_LdDPtkvZ7fsitv20h3p5yyzaLczF7T5fvL26xYplYLnhKSzUEScSe45tYYXythpddcAqJQRgNpm2snlcnzLVl0xtSo620NEgmn7PF468IhxkC--gpNZ8NYZVD9qVYn1V_y4Uha152h__EHwq1L2A</recordid><startdate>20250531</startdate><enddate>20250531</enddate><creator>Yang, Zhe</creator><creator>Wang, Qing</creator><creator>Liao, Xiaojian</creator><creator>Lu, Youyou</creator><creator>Huang, Keji</creator><creator>Shu, Jiwu</creator><general>ACM</general><scope>AAYXX</scope><scope>CITATION</scope><orcidid>https://orcid.org/0000-0002-6214-5390</orcidid><orcidid>https://orcid.org/0000-0002-2031-5937</orcidid><orcidid>https://orcid.org/0000-0002-7362-2789</orcidid><orcidid>https://orcid.org/0000-0002-5526-7154</orcidid><orcidid>https://orcid.org/0009-0009-5496-5522</orcidid><orcidid>https://orcid.org/0000-0002-7924-9268</orcidid></search><sort><creationdate>20250531</creationdate><title>Efficiently Enlarging RDMA-Attached Memory with SSD</title><author>Yang, Zhe ; Wang, Qing ; Liao, Xiaojian ; Lu, Youyou ; Huang, Keji ; Shu, Jiwu</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a842-e3ea605ee2c4282a99fd74a5f82503347980e8a68c57966bea3c99d38dbd053e3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2025</creationdate><topic>Distributed memory</topic><topic>Information storage systems</topic><topic>Information systems</topic><topic>Software and its engineering</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Yang, Zhe</creatorcontrib><creatorcontrib>Wang, Qing</creatorcontrib><creatorcontrib>Liao, Xiaojian</creatorcontrib><creatorcontrib>Lu, Youyou</creatorcontrib><creatorcontrib>Huang, Keji</creatorcontrib><creatorcontrib>Shu, Jiwu</creatorcontrib><collection>CrossRef</collection><jtitle>ACM transactions on storage</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Yang, Zhe</au><au>Wang, Qing</au><au>Liao, Xiaojian</au><au>Lu, Youyou</au><au>Huang, Keji</au><au>Shu, Jiwu</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Efficiently Enlarging RDMA-Attached Memory with SSD</atitle><jtitle>ACM transactions on storage</jtitle><stitle>ACM TOS</stitle><date>2025-05-31</date><risdate>2025</risdate><volume>21</volume><issue>2</issue><spage>1</spage><epage>27</epage><pages>1-27</pages><issn>1553-3077</issn><eissn>1553-3093</eissn><abstract>RDMA-based in-memory storage systems offer high performance but are restricted by the capacity of physical memory. In this paper, we propose TeRM to extend RDMA-attached memory with SSD. TeRM achieves fast remote access on the SSD-extended memory by eliminating page faults of RDMA NIC and CPU from the critical path. We also introduce a set of techniques to reduce the consumption of CPU and network resources. Evaluation shows that TeRM performs close to the performance of the ideal upper bound where all pages are pinned in the physical memory. Compared with existing approaches TeRM significantly improves the performance of unmodified RDMA-based storage systems, including a file system and a key-value system.</abstract><cop>New York, NY</cop><pub>ACM</pub><doi>10.1145/3700772</doi><tpages>27</tpages><orcidid>https://orcid.org/0000-0002-6214-5390</orcidid><orcidid>https://orcid.org/0000-0002-2031-5937</orcidid><orcidid>https://orcid.org/0000-0002-7362-2789</orcidid><orcidid>https://orcid.org/0000-0002-5526-7154</orcidid><orcidid>https://orcid.org/0009-0009-5496-5522</orcidid><orcidid>https://orcid.org/0000-0002-7924-9268</orcidid><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext |
| identifier | ISSN: 1553-3077 |
| ispartof | ACM transactions on storage, 2025-05, Vol.21 (2), p.1-27 |
| issn | 1553-3077 1553-3093 |
| language | eng |
| recordid | cdi_crossref_primary_10_1145_3700772 |
| source | ACM Digital Library Complete |
| subjects | Distributed memory Information storage systems Information systems Software and its engineering |
| title | Efficiently Enlarging RDMA-Attached Memory with SSD |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-21T05%3A44%3A27IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-acm_cross&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Efficiently%20Enlarging%20RDMA-Attached%20Memory%20with%20SSD&rft.jtitle=ACM%20transactions%20on%20storage&rft.au=Yang,%20Zhe&rft.date=2025-05-31&rft.volume=21&rft.issue=2&rft.spage=1&rft.epage=27&rft.pages=1-27&rft.issn=1553-3077&rft.eissn=1553-3093&rft_id=info:doi/10.1145/3700772&rft_dat=%3Cacm_cross%3E3700772%3C/acm_cross%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true |