[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-seo-341722-105":3,"doc-detail-341722-en":81,"detail-sidebar-cat-0-en-105":116},{"code":4,"msg":5,"data":6},0,"ok",{"site_id":7,"language":8,"slug":9,"title":10,"keywords":11,"description":12,"schema_data":13,"social_meta":74,"head_meta":76,"extra_data":78,"updated_unix":80},105,"en","camel-hash-table-striking-a-balance-between-cpu-and-memory-efficiency-in-main-memory-hash-join","CAMEL Hash Table - Striking a Balance Between CPU and Memory Efficiency in Main-Memory Hash Join","","Many hash tables used in hash joins consume substantial memory to maintain constant-time lookup. As modern workloads grow, hash joins become increasingly memory-intensive because probe operations depend on these tables, making memory-efficient designs essential for sustained query performance. The Concise Hash Table (CHT) was introduced as a space-efficient alternative, yet experiments show its performance degrades with larger data volumes and large results, limiting scalability in real scenarios.",{"@graph":14,"@context":73},[15,34,56],{"@type":16,"itemListElement":17},"BreadcrumbList",[18,23,27,31],{"item":19,"name":20,"@type":21,"position":22},"https://docshare.wps.com","Home","ListItem",1,{"item":24,"name":25,"@type":21,"position":26},"https://docshare.wps.com/document/","Document",2,{"item":28,"name":29,"@type":21,"position":30},"https://docshare.wps.com/document/research-report/","Research & Report",3,{"item":32,"name":10,"@type":21,"position":33},"https://docshare.wps.com/document/camel-hash-table-striking-a-balance-between-cpu-and-memory-efficiency-in-main-memory-hash-join/341722/",4,{"url":32,"name":10,"@type":35,"image":36,"author":41,"headline":10,"publisher":44,"fileFormat":47,"inLanguage":8,"description":12,"dateModified":48,"datePublished":49,"encodingFormat":47,"isAccessibleForFree":50,"interactionStatistic":51},"DigitalDocument",{"url":37,"@type":38,"width":39,"height":40},"https://docshare.wps.com/thumbnails/camel-hash-table-striking-a-balance-between-cpu-and-memory-efficiency-in-main-memory-hash-join/341722.png","ImageObject",300,407,{"name":42,"@type":43},"Valentina","Person",{"url":19,"name":45,"@type":46},"DocShare","Organization","application/pdf","2026-10-31","2026-09-22",true,{"@type":52,"interactionType":53,"userInteractionCount":55},"InteractionCounter",{"@type":54},"ViewAction",11,{"@type":57,"mainEntity":58},"FAQPage",[59,65,69],{"name":60,"@type":61,"acceptedAnswer":62},"Why do hash joins become memory-intensive in modern workloads?","Question",{"text":63,"@type":64},"Many hash tables used during hash joins require substantial memory to guarantee constant-time lookup, and probe operations rely on these tables. As data volume grows, this memory consumption increases.","Answer",{"name":66,"@type":61,"acceptedAnswer":67},"What limitation is observed for CHT-based hash joins?",{"text":68,"@type":64},"Experimental evaluation indicates that CHT-based hash joins degrade as data volumes and result sizes increase, which restricts scalability in real-world scenarios.",{"name":70,"@type":61,"acceptedAnswer":71},"How does CAMEL-HT improve performance and memory efficiency?",{"text":72,"@type":64},"CAMEL-HT is designed as a cache-aware, memory-efficient lightweight hash table. Experiments show it improves hash join performance by 2× over CHT with only a marginal increase in memory usage, and achieves about 3× better memory efficiency than VIP Hashing while often reducing execution time.","https://schema.org",{"og:url":32,"og:type":75,"og:title":10,"og:site_name":45,"og:description":12},"article",{"robots":77,"canonical":32},"index,follow",{"doc_id":79,"site_id":7},341722,1790508841,{"code":4,"msg":82,"data":83},"success",{"doc_id":79,"user_id":84,"nickname":42,"user_avatar":85,"doc_module":4,"category_id":86,"category_name":29,"doc_title":10,"doc_description":12,"doc_content":87,"file_id":88,"file_url":89,"file_type":90,"file_size":91,"view_count":55,"is_deleted":4,"is_public":22,"is_downloadable":22,"audit_status":22,"page_count":92,"language":93,"language_code":8,"site_id":7,"html_lang":8,"table_of_contents":94,"faqs":95,"seo_title":96,"seo_description":12,"update_tm":97,"read_time":98,"preview_page_count":33,"preview_urls":99},13056703020460,"https://ap-avatar.wpscdn.com/avatar/be000253dac470eee5d?_k=1778207105932848923",8,"␣  \nCAMEL Hash Table: Striking a Balance Between CPU and Memory Efficiency in Main-Memory Hash Join  \nSudip Chatterjee Xiaozheng Zhang Suprio Ray  \n[sudip.chatterjee@unb.ca](sudip.chatterjee@unb.ca)[ ](sudip.chatterjee@unb.ca)[xz.zhang@unb.ca](xz.zhang@unb.ca)[sray@unb.ca](sray@unb.ca)  \nUniversity of New Brunswick, Canada  \nIan Finlay Calisto Zuzarte Mark Stoodley  \n[finlay@ca.ibm.com](finlay@ca.ibm.com)[ ](finlay@ca.ibm.com)[calisto@ca.ibm.com](calisto@ca.ibm.com)[mstoodle@ca.ibm.com](mstoodle@ca.ibm.com)[ ](mstoodle@ca.ibm.com)IBM, Canada  \nAbstract  \nMany hash tables used in hash joins incur substantial memory consumption to guarantee constant-time lookup performance. Hash joins, which rely on these tables for probe operations, have become increasingly memory-intensive due to the growing volume of data in modern workloads. As a result, memory-efficient hash table designs are critical for maintaining overall query performance. To address this challenge, the Concise Hash Table (CHT) was introduced as a space-efficient alternative that supports hash joins while utilizing available memory more effectively. However, our experimental evaluation reveals that the performance ofCHT-based hash joins degrades with increasing data volumes and large results, limiting its scalability in realworld scenarios.  \nModern processors attempt to hide cache and translation lookaside buffer (TLB) misses by employing out-of-order execution. However, hash table lookups are highly unpredictable, which can result in severe cache miss rates and low performance. Moreover, cache sizes in modern processors have not grown commensurately with increasing data volumes, making data prefetching and efficient cache utilization increasingly important. In this work, we propose a novel hash table, CAMEL-HT: a Cache Aware, Memory Efficient, Lightweight hash table that delivers high performance while consuming significantly less memory. Our experiments demonstrate that CAMEL-HT improves hash join performance by 2× over CHT with only a marginal increase in memory usage. CAMEL-HT consistently achieves 3× better memory efficiency compared to VIP Hashing, a state-ofthe-art pointer chasing hash table, while also outperforming it in several scenarios in terms of execution time. CAMEL-HT matches the performance of the state-of-the-art hash table designs while achieving memory efficiency comparable to the most compact hash table approaches.  \nKeywords  \nHash Join, In-Memory, Cache Aware, Memory Efficient, Compact Hash Table, CAMEL-HT  \n1 Introduction  \nHash tables [11] are foundational data structures designed to efficiently manage key-value pairs, offering constant-time lookups, low computational overhead, and favorable memory access patterns. These properties make them particularly well-suited for high-performance data processing in modern database systems.  \nEDBT’26, Tampere (Finland)  \n© 2026 Copyright held by the owner/author(s). [Published on OpenProceedings.org](Published on OpenProceedings.org)[ ](Published on OpenProceedings.org)[under ISBN 978-3-89318-104-9](under ISBN 978-3-89318-104-9), series ISSN 2367-2005. Distribution of this paper is permitted under the terms of the Creative Commons license CC-by-nc-nd 4.0.  \nFigure 1: Normalized execution time and memory usage (min–max scaling) for Concise Hash Table (CHT), CAMEL-HT, and VIP Hashing on the TPC-H dataset (scale factor 10)  \nThey are particularly relevant for in-memory indexing [29, 37], hash joins [1, 3, 4, 7, 10, 14, 16, 31], and aggregation operations [24, 27], where fast key-based access and efficient handling of large datasets are essential for performance.  \nThe performance of the hash-join operator is heavily influenced by the underlying hash table design. Numerous hash-table architectures have been proposed over the years to optimize this operation. In general, these designs can be categorized based on their collision resolution strategies into two primary classes: separate chaining [11] and open add","cbCaicQ2WJISuO5U","https://ap.wps.com/l/cbCaicQ2WJISuO5U","pdf",1271853,13,"English","# Abstract\n# Keywords\n# 1 Introduction","[{\"question\":\"Why do hash joins become memory-intensive in modern workloads?\",\"answer\":\"Many hash tables used during hash joins require substantial memory to guarantee constant-time lookup, and probe operations rely on these tables. As data volume grows, this memory consumption increases.\"},{\"question\":\"What limitation is observed for CHT-based hash joins?\",\"answer\":\"Experimental evaluation indicates that CHT-based hash joins degrade as data volumes and result sizes increase, which restricts scalability in real-world scenarios.\"},{\"question\":\"How does CAMEL-HT improve performance and memory efficiency?\",\"answer\":\"CAMEL-HT is designed as a cache-aware, memory-efficient lightweight hash table. Experiments show it improves hash join performance by 2× over CHT with only a marginal increase in memory usage, and achieves about 3× better memory efficiency than VIP Hashing while often reducing execution time.\"}]","CAMEL Hash Table - Striking a Balance Between CPU and Memory Efficiency in Main-Memory Hash Join | PDF",1790043360,33,[100,104,108,112],{"page_no":22,"url":101,"width":102,"height":103},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/12cd6ec3712814aab9c7070f183ead6c.webp",983,1390,{"page_no":26,"url":105,"width":106,"height":107},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/dae724ca36b6cd2ec0b5a0a8ffea6162.webp",982,1388,{"page_no":30,"url":109,"width":110,"height":111},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/101be51fd763c8ff11e84922e0e76748.webp",981,1387,{"page_no":33,"url":113,"width":114,"height":115},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/a95632a679911f28c442a2e1f8841674.webp",980,1385,{"code":4,"msg":82,"data":117},[118,122,126,130,135,140,145,148,153,156,160],{"id":22,"doc_module":4,"doc_module_name":25,"category_name":119,"show_sort_weight":120,"slug":121},"Story & Novel",90,"story-novel",{"id":26,"doc_module":4,"doc_module_name":25,"category_name":123,"show_sort_weight":124,"slug":125},"Literature",80,"literature",{"id":33,"doc_module":4,"doc_module_name":25,"category_name":127,"show_sort_weight":128,"slug":129},"Exam",70,"exam",{"id":131,"doc_module":4,"doc_module_name":25,"category_name":132,"show_sort_weight":133,"slug":134},5,"Comic",60,"comic",{"id":136,"doc_module":4,"doc_module_name":25,"category_name":137,"show_sort_weight":138,"slug":139},6,"Technology",50,"technology",{"id":141,"doc_module":4,"doc_module_name":25,"category_name":142,"show_sort_weight":143,"slug":144},7,"Healthcare",40,"healthcare",{"id":86,"doc_module":4,"doc_module_name":25,"category_name":29,"show_sort_weight":146,"slug":147},30,"research-report",{"id":149,"doc_module":4,"doc_module_name":25,"category_name":150,"show_sort_weight":151,"slug":152},9,"Religion & Spirituality",20,"religion-spirituality",{"id":151,"doc_module":4,"doc_module_name":25,"category_name":154,"show_sort_weight":151,"slug":155},"World Cup","world-cup",{"id":157,"doc_module":4,"doc_module_name":25,"category_name":158,"show_sort_weight":157,"slug":159},10,"Lifestyle","lifestyle",{"id":161,"doc_module":4,"doc_module_name":25,"category_name":162,"show_sort_weight":131,"slug":163},19,"General","general"]