{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,9,21]],"date-time":"2025-09-21T07:15:15Z","timestamp":1758438915017,"version":"3.44.0"},"reference-count":56,"publisher":"Association for Computing Machinery (ACM)","issue":"3","funder":[{"DOI":"10.13039\/501100001809","name":"National Science Foundation of China","doi-asserted-by":"crossref","award":["62272252, 62272253"],"award-info":[{"award-number":["62272252, 62272253"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["ACM Trans. Archit. Code Optim."],"published-print":{"date-parts":[[2025,9,30]]},"abstract":"<jats:p>Learned indexes provide significant performance advantages over classical ordered indexes. However, current learned indexes face challenges regarding tradeoffs between performance and space, as well as scalability issues in platforms with multiple NUMA nodes. These limitations hinder the practical application of learned indexes in production environments. This article presents DiffLex, a learned index with high-performance, memory-efficiency, and NUMA-awareness. The core design of DiffLex is to perform differentiated management based on the popularity of data. For optimal performance, DiffLex stores newly inserted data in sparse delta arrays and frequently accessed data in sparse hot cache arrays. However, for cold data that occupy a majority of the storage space, DiffLex stores them in dense arrays and conducts compression to reduce memory costs. DiffLex ensures NUMA-awareness by partitioning sparse deltas and replicating the hot cache arrays across multiple NUMA nodes. Additionally, we propose a persistent version of DiffLex tailored for emerging persistent memory devices. Our evaluation results demonstrate that DiffLex achieving 3.88\u00d7 and 1.82\u00d7 performance improvements compared to state-of-the-art learned indexes, while maintaining a compact index size.<\/jats:p>","DOI":"10.1145\/3736168","type":"journal-article","created":{"date-parts":[[2025,6,12]],"date-time":"2025-06-12T07:32:24Z","timestamp":1749713544000},"page":"1-26","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Towards Optimizing Learned Index for High Performance, Memory Efficiency and NUMA Awareness"],"prefix":"10.1145","volume":"22","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-4017-0974","authenticated-orcid":false,"given":"Lixiao","family":"Cui","sequence":"first","affiliation":[{"name":"College of Computer Science, Nankai University","place":["Tianjin, China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0004-9770-9987","authenticated-orcid":false,"given":"Kedi","family":"Yang","sequence":"additional","affiliation":[{"name":"College of Computer Science, Nankai University","place":["Tianjin, China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6623-350X","authenticated-orcid":false,"given":"Yusen","family":"Li","sequence":"additional","affiliation":[{"name":"College of Computer Science, Nankai University","place":["Tianjin, China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0387-2501","authenticated-orcid":false,"given":"Gang","family":"Wang","sequence":"additional","affiliation":[{"name":"College of Computer Science, Nankai University","place":["Tianjin, China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9010-3278","authenticated-orcid":false,"given":"Xiaoguang","family":"Liu","sequence":"additional","affiliation":[{"name":"College of Computer Science, Nankai University","place":["Tianjin, China"]}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"320","published-online":{"date-parts":[[2025,9,19]]},"reference":[{"key":"e_1_3_1_2_2","doi-asserted-by":"publisher","DOI":"10.1145\/3514221.3526121"},{"key":"e_1_3_1_3_2","doi-asserted-by":"publisher","DOI":"10.1007\/BF00263762"},{"key":"e_1_3_1_4_2","doi-asserted-by":"publisher","DOI":"10.1145\/376284.375681"},{"key":"e_1_3_1_5_2","doi-asserted-by":"publisher","DOI":"10.1145\/3037697.3037721"},{"key":"e_1_3_1_6_2","doi-asserted-by":"publisher","DOI":"10.5555\/3386691.3386712"},{"key":"e_1_3_1_7_2","doi-asserted-by":"publisher","DOI":"10.1145\/2660193.2660224"},{"key":"e_1_3_1_8_2","first-page":"239","volume-title":"Proceedings of the 18th USENIX Conference on File and Storage Technologies","author":"Chen Jiqiang","year":"2020","unstructured":"Jiqiang Chen, Liang Chen, Sheng Wang, Guoyun Zhu, Yuanyuan Sun, Huan Liu, and Feifei Li. 2020. HotRing: A hotspot-aware in-memory key-value store. In Proceedings of the 18th USENIX Conference on File and Storage Technologies. 239\u2013252."},{"key":"e_1_3_1_9_2","doi-asserted-by":"publisher","DOI":"10.1109\/ICDEW53142.2021.00019"},{"key":"e_1_3_1_10_2","doi-asserted-by":"publisher","DOI":"10.14778\/2752939.2752947"},{"key":"e_1_3_1_11_2","first-page":"321","volume-title":"Proceedings of the 2017 USENIX Annual Technical Conference","author":"Cidon Asaf","year":"2017","unstructured":"Asaf Cidon, Daniel Rushton, Stephen M. Rumble, and Ryan Stutsman. 2017. Memshare: A dynamic multi-tenant key-value cache. In Proceedings of the 2017 USENIX Annual Technical Conference. USENIX Association, Santa Clara, CA, 321\u2013334."},{"key":"e_1_3_1_12_2","doi-asserted-by":"publisher","DOI":"10.1145\/1807128.1807152"},{"key":"e_1_3_1_13_2","article-title":"Intel Optane DC Persistent Memory","author":"Corporation Intel","year":"2022","unstructured":"Intel Corporation. 2022. Intel Optane DC Persistent Memory. Retrieved 25 March 2024 from https:\/\/www.intel.com\/content\/www\/us\/en\/products%\/docs\/memory-storage\/optane-persistent-memory\/overview.html","journal-title":"R"},{"key":"e_1_3_1_14_2","doi-asserted-by":"publisher","DOI":"10.1109\/TKDE.2023.3342825"},{"key":"e_1_3_1_15_2","doi-asserted-by":"publisher","DOI":"10.1145\/3605573.3605590"},{"key":"e_1_3_1_16_2","volume-title":"Proceedings of the 32nd International Symposium on Distributed Computing.","author":"Daly Henry","year":"2018","unstructured":"Henry Daly, Ahmed Hassan, Michael F Spear, and Roberto Palmieri. 2018. NUMASK: High performance scalable skip list for NUMA. In Proceedings of the 32nd International Symposium on Distributed Computing. Schloss Dagstuhl-Leibniz-Zentrum fuer Informatik."},{"key":"e_1_3_1_17_2","doi-asserted-by":"publisher","DOI":"10.1145\/3318464.3389711"},{"key":"e_1_3_1_18_2","doi-asserted-by":"publisher","DOI":"10.14778\/3389133.3389135"},{"key":"e_1_3_1_19_2","doi-asserted-by":"publisher","DOI":"10.1145\/3299869.3319860"},{"key":"e_1_3_1_20_2","doi-asserted-by":"publisher","DOI":"10.1145\/3654972"},{"key":"e_1_3_1_21_2","unstructured":"Google. 2024. Snappy. Retrieved 8 April 2024 from https:\/\/github.com\/google\/snappy"},{"key":"e_1_3_1_22_2","doi-asserted-by":"publisher","DOI":"10.5555\/3189759.3189777"},{"key":"e_1_3_1_23_2","doi-asserted-by":"publisher","DOI":"10.1109\/TPDS.2019.2945315"},{"key":"e_1_3_1_24_2","doi-asserted-by":"publisher","DOI":"10.14778\/3547305.3547306"},{"key":"e_1_3_1_25_2","doi-asserted-by":"publisher","DOI":"10.1145\/3401071.3401659"},{"key":"e_1_3_1_26_2","doi-asserted-by":"publisher","DOI":"10.1145\/3183713.3196909"},{"key":"e_1_3_1_27_2","first-page":"773","volume-title":"Proceedings of the 2021 USENIX Annual Technical Conference","author":"Krishnan R. Madhava","year":"2021","unstructured":"R. Madhava Krishnan, Wook-Hee Kim, Xinwei Fu, Sumit Kumar Monga, Hee Won Lee, Minsung Jang, Ajit Mathew, and Changwoo Min. 2021. TIPS: Making volatile index structures persistent with DRAM-NVMM tiering. In Proceedings of the 2021 USENIX Annual Technical Conference. USENIX Association, 773\u2013787."},{"key":"e_1_3_1_28_2","first-page":"257","volume-title":"Proceedings of the 15th USENIX Conference on File and Storage Technologies","author":"Lee Se Kwon","year":"2017","unstructured":"Se Kwon Lee, K Hyun Lim, Hyunsub Song, Beomseok Nam, and Sam H Noh. 2017. WORT: Write optimal radix tree for persistent memory storage systems. In Proceedings of the 15th USENIX Conference on File and Storage Technologies. 257\u2013270."},{"key":"e_1_3_1_29_2","doi-asserted-by":"publisher","DOI":"10.1145\/2933349.2933352"},{"key":"e_1_3_1_30_2","doi-asserted-by":"publisher","DOI":"10.14778\/3489496.3489512"},{"key":"e_1_3_1_31_2","doi-asserted-by":"publisher","DOI":"10.14778\/3384345.3384355"},{"key":"e_1_3_1_32_2","doi-asserted-by":"publisher","DOI":"10.14778\/3494124.3494141"},{"key":"e_1_3_1_33_2","doi-asserted-by":"publisher","DOI":"10.14778\/3389133.3389134"},{"key":"e_1_3_1_34_2","first-page":"1","volume-title":"Proceedings of the 19th USENIX Conference on File and Storage Technologies","author":"Ma Shaonan","year":"2021","unstructured":"Shaonan Ma, Kang Chen, Shimin Chen, Mengxing Liu, Jianglang Zhu, Hongbo Kang, and Yongwei Wu. 2021. ROART: Range-query optimized persistent ART. In Proceedings of the 19th USENIX Conference on File and Storage Technologies. 1\u201316."},{"key":"e_1_3_1_35_2","doi-asserted-by":"publisher","DOI":"10.14778\/3397230.3397232"},{"key":"e_1_3_1_36_2","doi-asserted-by":"publisher","DOI":"10.1145\/3373376.3378456"},{"key":"e_1_3_1_37_2","doi-asserted-by":"publisher","DOI":"10.5555\/3323298.3323302"},{"key":"e_1_3_1_38_2","doi-asserted-by":"publisher","DOI":"10.1145\/3318464.3380579"},{"key":"e_1_3_1_39_2","doi-asserted-by":"publisher","DOI":"10.1145\/2882903.2915251"},{"key":"e_1_3_1_40_2","article-title":"LZ4","author":"Pavlov Igor","year":"2024","unstructured":"Igor Pavlov. 2024. LZ4. Retrieved 8 April 2024 from https:\/\/github.com\/lz4\/lz4","journal-title":"R"},{"key":"e_1_3_1_41_2","article-title":"LZMA","author":"Pavlov Igor","year":"2024","unstructured":"Igor Pavlov. 2024. LZMA. Retrieved 8 April 2024 from https:\/\/www.7-zip.org\/sdk.html","journal-title":"R"},{"key":"e_1_3_1_42_2","doi-asserted-by":"publisher","DOI":"10.1147\/rd.524.0465"},{"key":"e_1_3_1_43_2","doi-asserted-by":"publisher","DOI":"10.14778\/3594512.3594528"},{"key":"e_1_3_1_44_2","doi-asserted-by":"publisher","DOI":"10.1145\/3332466.3374547"},{"key":"e_1_3_1_45_2","doi-asserted-by":"publisher","DOI":"10.1145\/1950365.1950379"},{"key":"e_1_3_1_46_2","doi-asserted-by":"publisher","DOI":"10.14778\/3397230.3397247"},{"key":"e_1_3_1_47_2","doi-asserted-by":"publisher","DOI":"10.1145\/3409963.3410496"},{"key":"e_1_3_1_48_2","doi-asserted-by":"publisher","DOI":"10.14778\/3551793.3551848"},{"key":"e_1_3_1_49_2","doi-asserted-by":"publisher","DOI":"10.14778\/3457390.3457393"},{"key":"e_1_3_1_50_2","doi-asserted-by":"publisher","DOI":"10.14778\/3547305.3547322"},{"key":"e_1_3_1_51_2","doi-asserted-by":"publisher","DOI":"10.1109\/HPCA.2015.7056056"},{"key":"e_1_3_1_52_2","doi-asserted-by":"publisher","DOI":"10.5555\/3386691.3386708"},{"key":"e_1_3_1_53_2","first-page":"167","volume-title":"Proceedings of the 13th USENIX Conference on File and Storage Technologies","author":"Yang Jun","year":"2015","unstructured":"Jun Yang, Qingsong Wei, Cheng Chen, Chundong Wang, Khai Leong Yong, and Bingsheng He. 2015. NV-Tree: Reducing consistency cost for NVM-based single level systems. In Proceedings of the 13th USENIX Conference on File and Storage Technologies. Santa Clara, CA, 167\u2013181."},{"key":"e_1_3_1_54_2","doi-asserted-by":"publisher","DOI":"10.1145\/3468521"},{"key":"e_1_3_1_55_2","doi-asserted-by":"publisher","DOI":"10.1145\/2882903.2915222"},{"key":"e_1_3_1_56_2","doi-asserted-by":"publisher","DOI":"10.14778\/3372716.3372717"},{"key":"e_1_3_1_57_2","first-page":"461","volume-title":"Proceedings of the 13th USENIX Symposium on Operating System Design and Implementation","author":"Zuo Pengfei","year":"2018","unstructured":"Pengfei Zuo, Yu Hua, and Jie Wu. 2018. Write-optimized and high-performance hashing index scheme for persistent memory. In Proceedings of the 13th USENIX Symposium on Operating System Design and Implementation. 461\u2013476."}],"container-title":["ACM Transactions on Architecture and Code Optimization"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3736168","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,9,20]],"date-time":"2025-09-20T00:49:44Z","timestamp":1758329384000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3736168"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,9,19]]},"references-count":56,"journal-issue":{"issue":"3","published-print":{"date-parts":[[2025,9,30]]}},"alternative-id":["10.1145\/3736168"],"URL":"https:\/\/doi.org\/10.1145\/3736168","relation":{},"ISSN":["1544-3566","1544-3973"],"issn-type":[{"type":"print","value":"1544-3566"},{"type":"electronic","value":"1544-3973"}],"subject":[],"published":{"date-parts":[[2025,9,19]]},"assertion":[{"value":"2024-04-11","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-04-28","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-09-19","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}