DBScholar

Back to papers

The Case for Learned In-Memory Joins

Summary: Explores CDF-based learned indexes (RMI, RadixSpline) for in-memory indexed nested-loop, sort, and hash joins. Optimized learned variants outperform state-of-the-art joins across diverse datasets and workloads. (summarized by gpt-5.6-luna on Jul 24 2026)

Paper ID
13221
Venue
VLDB
Year
2023
Pagerank
5.460955e-05
Overall Rank
8,234 | 43.51%
DOI
10.14778/3587136.3587148

Incoming Non-self Citations Over Time

Authors

BibTeX Citation

@article{sabek_vldb23,
        title = {{The Case for Learned In-Memory Joins}},
        author = {Sabek, Ibrahim and Kraska, Tim},
        journal = {PVLDB},
        series = {{VLDB} '23},
        volume = {16},
        number = {7},
        pages = {1749--1762},
        doi = {10.14778/3587136.3587148},
        url = {https://doi.org/10.14778/3587136.3587148},
        year = {2023}
}

Incoming Citations (Sorted by Pagerank)

Showing 5 of 5 citing papers.

Rank Citing Paper Year Venue Pagerank
7,846 The Holon Approach for Simultaneously Tuning Multiple Components in a Self-Driving Database Management System with Machine Learning via Synthesized Proto-Actions 2024 VLDB 5.5331459e-05
9,465 Are Joins over LSM-trees Ready? Take RocksDB as an Example 2025 VLDB 5.2634238e-05
10,506 This is Going to Sound Crazy, But What If We Used Large Language Models to Boost Automatic Database Tuning Algorithms By Leveraging Prior History? We Will Find Better Configurations More Quickly Than Retraining From Scratch! 2026 SIGMOD 5.093636e-05
10,955 DobLIX: A Dual-Objective Learned Index for Log-Structured Merge Trees 2025 VLDB 5.093636e-05
11,204 SPID-Join: A Skew-resistant Processing-in-DIMM Join Algorithm Exploiting the Bank- and Rank-level Parallelisms of DIMMs 2024 SIGMOD 5.093636e-05
Previous Page 1 / 1 Next

Outgoing Citations (Sorted by Pagerank)

Showing 35 of 35 cited papers.

Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.

Rank Cited Paper Year Venue Pagerank
18 How Good Are Query Optimizers, Really? 2016 VLDB 0.00059284255
29 Database Architecture Optimized for the New Bottleneck: Memory Access 1999 VLDB 0.00052093615
43 The Case for Learned Index Structures 2018 SIGMOD 0.00046060254
84 Learned Cardinalities: Estimating Correlated Joins with Deep Learning 2019 CIDR 0.00035838391
154 Neo: A Learned Query Optimizer 2019 VLDB 0.00028726181
209 Sort vs. Hash Revisited: Fast Join Implementation on Modern Multi-Core CPUs 2009 VLDB 0.00024932174
252 Multi-Core, Main-Memory Joins: Sort vs. Hash Revisited 2014 VLDB 0.00023242719
360 Design and Evaluation of Main Memory Hash Join Algorithms for Multi-core CPUs 2011 SIGMOD 0.00020182846
378 Bao: Making Learned Query Optimization Practical 2021 SIGMOD 0.00019638121
422 Umbra: A Disk-Based System with In-Memory Performance 2020 CIDR 0.00018732744
423 Massively Parallel Sort-Merge Joins in Main Memory Multi-Core Database Systems 2012 VLDB 0.00018725853
447 ALEX: An Updatable Adaptive Learned Index 2020 SIGMOD 0.00018322593
477 The PGM-index: a fully-dynamic compressed learned index with provable worst-case bounds 2020 VLDB 0.00017851226
631 Relational Joins on Graphics Processors 2008 SIGMOD 0.00015591241
847 Benchmarking Learned Indexes 2021 VLDB 0.0001365768
873 Learning Multi-dimensional Indexes 2020 SIGMOD 0.00013481915
987 What happens during a Join? Dissecting CPU and Memory Optimization Effects 2000 VLDB 0.00012814017
1,110 Buffering Accesses to Memory-Resident Index Structures 2003 VLDB 0.00012140661
1,135 Qd-tree: Learning Data Layouts for Big Data Analytics 2020 SIGMOD 0.00012032847
1,203 NUMA-aware algorithms: the case of data shuffling 2013 CIDR 0.00011671169
1,265 An Experimental Comparison of Thirteen Relational Equi-Joins in Main Memory 2016 SIGMOD 0.00011415709
1,551 Updatable Learned Index with Precise Positions 2021 VLDB 0.00010381398
2,156 Quickstep: A Data Platform Based on the Scaling-Up Approach 2018 VLDB 9.0635624e-05
2,649 Asynchronous Memory Access Chaining 2016 VLDB 8.2926258e-05
2,767 Exploiting Coroutines to Attack the "Killer Nanoseconds" 2018 VLDB 8.1507611e-05
2,806 Are Updatable Learned Indexes Ready? 2022 VLDB 8.1013097e-05
2,926 Distributed Join Algorithms on Thousands of Cores 2017 VLDB 7.9549783e-05
2,962 To Partition, or Not to Partition, That is the Join Question in a Real System 2021 SIGMOD 7.9170451e-05
3,865 The Case for a Learned Sorting Algorithm 2020 SIGMOD 7.0621718e-05
4,780 Can Learned Models Replace Hash Functions? 2023 VLDB 6.5118885e-05
5,148 LSched: A Workload-Aware Learned Query Scheduler for Analytical Database Systems 2022 SIGMOD 6.3465986e-05
5,974 Towards instance-optimized data systems 2021 VLDB 6.0230488e-05
6,163 A Critical Analysis of Recursive Model Indexes 2022 VLDB 5.9532723e-05
7,792 Cache-Oblivious Query Processing 2007 CIDR 5.5431413e-05
9,939 Is FPGA Useful for Hash Joins? Exploring Hash Joins on Coupled CPU-FPGA Architecture 2020 CIDR 5.1947648e-05
Previous Page 1 / 1 Next

Semantically Similar Papers

# Overall Rank Paper Year Venue
1 3,011 Design and Evaluation of Parallel Pipelined Join Algorithms 1987 SIGMOD
2 4,780 Can Learned Models Replace Hash Functions? 2023 VLDB
3 252 Multi-Core, Main-Memory Joins: Sort vs. Hash Revisited 2014 VLDB
4 3,255 Processing Sliding Window Multi-Joins in Continuous Queries over Data Streams 2003 VLDB
5 8,468 Making RDBMSs Efficient on Graph Workloads Through Predefined Joins 2022 VLDB
6 5,812 MCJoin: A Memory-Constrained Join for Column-Store Main-Memory Databases. 2012 SIGMOD
7 1,740 Adopting Worst-Case Optimal Joins in Relational Database Systems 2020 VLDB
8 360 Design and Evaluation of Main Memory Hash Join Algorithms for Multi-core CPUs 2011 SIGMOD
9 1,265 An Experimental Comparison of Thirteen Relational Equi-Joins in Main Memory 2016 SIGMOD
10 959 Memory-Efficient Hash Joins 2015 VLDB