DBScholar

Back to papers

On Optimizing Operator Fusion Plans for Large-Scale Machine Learning in SystemML

Summary: Exact, cost-based fusion optimization for large-scale ML in SystemML; handles complex operator DAGs and hybrid local/distributed execution. Integrated with candidate exploration and code generation for dense, sparse, and compressed data; up to 22x speedups with negligible compilation overhead. (summarized by gpt-5-nano on Feb 09 2026)

Paper ID
h08c62adea104db11
Venue
VLDB
Year
2018
Pagerank
7.6456038e-05
Overall Rank
3,103 | 79.15%
DOI
10.14778/3229863.3229865
PDF
Download (CC BY-NC-ND 4.0)

Incoming Non-self Citations Over Time

Authors

BibTeX Citation

@article{boehm_vldb18,
        title = {{On Optimizing Operator Fusion Plans for Large-Scale Machine Learning in SystemML}},
        author = {Boehm, Matthias and Reinwald, Berthold and Hutchison, Dylan and Sen, Prithviraj and Evfimievski, Alexandre V. and Pansare, Niketan},
        journal = {PVLDB},
        series = {{VLDB} '18},
        volume = {11},
        number = {12},
        pages = {1755--1768},
        doi = {10.14778/3229863.3229865},
        url = {https://doi.org/10.14778/3229863.3229865},
        year = {2018}
}

Incoming Citations (Sorted by Pagerank)

Showing 25 of 25 citing papers.

Rank Citing Paper Year Venue Pagerank
1,669 SystemDS: A Declarative Machine Learning System for the End-to-End Data Science Lifecycle 2020 CIDR 9.9324573e-05
2,266 An Intermediate Representation for Optimizing Machine Learning Pipelines 2019 VLDB 8.7248802e-05
4,097 Distributed Deep Learning on Data Systems: A Comparative Analysis of Approaches 2021 VLDB 6.8064316e-05
4,330 MNC: Structure-Exploiting Sparsity Estimation for Matrix Expressions 2019 SIGMOD 6.6564176e-05
4,334 LIMA: Fine-grained Lineage Tracing and Reuse in Machine Learning Systems 2021 SIGMOD 6.6537801e-05
4,975 SPORES: Sum-Product Optimization via Relational Equality Saturation for Large Scale Linear Algebra 2020 VLDB 6.3289021e-05
5,444 Optimizing Tensor Programs on Flexible Storage 2023 SIGMOD 6.1250455e-05
6,666 UPLIFT: Parallelization Strategies for Feature Transformations in Machine Learning Workloads 2022 VLDB 5.7144587e-05
6,883 DAPHNE: An Open and Extensible System Infrastructure for Integrated Data Analysis Pipelines 2022 CIDR 5.6552024e-05
7,988 Efficient Execution of User-Defined Functions in SQL Queries 2023 VLDB 5.4105672e-05
8,389 AWARE: Workload-aware, Redundancy-exploiting Linear Algebra 2023 SIGMOD 5.3396465e-05
8,405 TOD: GPU-accelerated Outlier Detection via Tensor Operations 2023 VLDB 5.3370629e-05
8,528 FuseME: Distributed Matrix Computation Engine based on Cuboid-based Fused Operator and Plan Generation 2022 SIGMOD 5.3212942e-05
8,777 PreVision: An Out-of-Core Matrix Computation System with Optimal Buffer Replacement 2024 SIGMOD 5.2789044e-05
9,166 HADAD: A Lightweight Approach for Optimizing Hybrid Complex Analytics Queries 2021 SIGMOD 5.2144986e-05
9,545 GIO: Generating Efficient Matrix and Frame Readers for Custom Data Formats by Example 2023 SIGMOD 5.1600923e-05
9,564 Towards an Optimized GROUP BY Abstraction for Large-Scale Machine Learning 2021 VLDB 5.1547835e-05
9,663 BladeDISC: Optimizing Dynamic Shape Machine Learning Workloads via Compiler Approach 2023 SIGMOD 5.142891e-05
9,700 The UDFBench Benchmark for General-purpose UDF Queries 2025 VLDB 5.1372261e-05
10,459 EncoderForge: Generating Efficient SQL for Encoders in Machine Learning Inference Pipelines 2026 SIGMOD 4.9769913e-05
10,867 Sema: A High-performance System for LLM-based Semantic Query Processing 2026 VLDB 4.9769913e-05
10,907 stratum: A System Infrastructure for Massive Agent-Centric ML Workloads 2026 VLDB 4.9769913e-05
11,251 Quantum Data Management in the NISQ Era 2025 VLDB 4.9769913e-05
11,852 Redundancy Elimination in Distributed Matrix Computation 2022 SIGMOD 4.9769913e-05
11,982 Hybrid Evaluation for Distributed Iterative Matrix Computation 2021 SIGMOD 4.9769913e-05
Previous Page 1 / 1 Next

Outgoing Citations (Sorted by Pagerank)

Showing 41 of 41 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 MAGIC SETS AND OTHER STRANGE WAYS TO IMPLEMENT LOGIC PROGRAMS (Extended Abstract) 1986 PODS 0.00058997063
21 Efficiently Compiling Efficient Query Plans for Modern Hardware 2011 VLDB 0.00056835296
23 Spark SQL: Relational Data Processing in Spark 2015 SIGMOD 0.00055384955
91 On the Propagation of Errors in the Size of Join Results 1991 SIGMOD 0.00034748721
154 MAD Skills: New Analysis Practices for Big Data 2009 VLDB 0.00028568843
224 Self-Driving Database Management Systems 2017 CIDR 0.00024011047
416 SystemML: Declarative Machine Learning on Spark 2016 VLDB 0.00018650998
495 Building Efficient Query Engines in a High-Level Language 2014 VLDB 0.00017363171
883 Dynamic Programming Strikes Back 2008 SIGMOD 0.00013263866
907 Data Blocks: Hybrid OLTP and OLAP on Compressed Storage using both Vectorization and Compilation 2016 SIGMOD 0.00013157412
1,034 A Scalable, Predictable Join Operator for Highly Concurrent Data Warehouses 2009 VLDB 0.00012389548
1,040 The DataPath System: A Data-Centric Analytic Processing Engine for Large Data Warehouses 2010 SIGMOD 0.00012358804
1,082 Hybrid Parallelization Strategies for Large-Scale Machine Learning in SystemML 2014 VLDB 0.00012118261
1,132 Efficient Exploitation of Similar Subexpressions for Query Processing 2007 SIGMOD 0.00011893781
1,185 Adaptive Optimization of Very Large Join Queries 2018 SIGMOD 0.00011607329
1,193 Weld: A Common Runtime for High Performance Data Analytics 2017 CIDR 0.00011582619
1,301 Analysis of Two Existing and One New Dynamic Programming Algorithm for the Generation of Optimal Bushy Join Trees without Cross Products 2006 VLDB 0.00011107788
1,341 Relaxed Operator Fusion for In-Memory Databases: Making Compilation, Vectorization, and Prefetching Work Together At Last 2018 VLDB 0.00010963427
1,445 An Architecture for Compiling UDF-centric Workflows 2015 VLDB 0.00010628379
1,473 Voodoo - A Vector Algebra for Portable Database Performance on Modern Hardware 2016 VLDB 0.00010552993
1,534 Pipelined Query Processing in Coprocessor Environments 2018 SIGMOD 0.00010327147
1,548 Conditioning Probabilistic Databases 2008 VLDB 0.00010288842
1,614 Compressed Linear Algebra for Large-Scale Machine Learning 2016 VLDB 0.00010067153
1,744 How to Architect a Query Compiler 2016 SIGMOD 9.7332903e-05
2,029 Scalable Join Processing on Very Large RDF Graphs 2009 SIGMOD 9.1580838e-05
2,125 Tupleware: "Big" Data, Big Analytics, Small Clusters 2015 CIDR 9.0017828e-05
2,225 Shared Workload Optimization 2014 VLDB 8.8062552e-05
2,249 Cumulon: Optimizing Statistical Data Analysis in the Cloud 2013 SIGMOD 8.7544468e-05
2,527 Stubby: A Transformation-based Optimizer for MapReduce Workflows 2012 VLDB 8.3386546e-05
2,724 Exploiting Matrix Dependency for Efficient Distributed Matrix Computation 2015 SIGMOD 8.090674e-05
2,753 Implicit Parallelism through Deep Language Embedding 2015 SIGMOD 8.0513412e-05
3,041 Configuration-Parametric Query Optimization for Physical Design Tuning 2008 SIGMOD 7.718081e-05
3,215 Efficient and Provable Multi-Query Optimization 2017 PODS 7.5233633e-05
3,331 SPOOF: Sum-Product Optimization and Operator Fusion for Large-Scale Machine Learning 2017 CIDR 7.4138851e-05
3,505 Fast Queries Over Heterogeneous Data Through Engine Customization 2016 VLDB 7.2474175e-05
3,722 Bypassing Joins in Disjunctive Queries 1995 VLDB 7.0684532e-05
3,800 Query Simplification: Graceful Degradation for Join-Order Optimization 2009 SIGMOD 7.0130412e-05
4,117 Resource Elasticity for Large-Scale Machine Learning 2015 SIGMOD 6.7929814e-05
4,900 Micro-architectural Analysis of In-memory OLTP 2016 SIGMOD 6.3625602e-05
7,780 DBToaster: Agile Views in a Dynamic Data Management System 2011 CIDR 5.4528185e-05
7,787 Measuring and Optimizing Distributed Array Programs 2016 VLDB 5.4511698e-05
Previous Page 1 / 1 Next

Semantically Similar Papers

# Overall Rank Paper Year Venue
1 9,829 BlockJoin: Efficient Matrix Partitioning Through Joins 2017 VLDB
2 6,146 Efficient Construction of Approximate Ad-Hoc ML models Through Materialization and Reuse 2018 VLDB
3 9,564 Towards an Optimized GROUP BY Abstraction for Large-Scale Machine Learning 2021 VLDB
4 8,528 FuseME: Distributed Matrix Computation Engine based on Cuboid-based Fused Operator and Plan Generation 2022 SIGMOD
5 6,171 Automatic Optimization of Matrix Implementations for Distributed Machine Learning and Linear Algebra 2021 SIGMOD
6 4,117 Resource Elasticity for Large-Scale Machine Learning 2015 SIGMOD
7 4,975 SPORES: Sum-Product Optimization via Relational Equality Saturation for Large Scale Linear Algebra 2020 VLDB
8 3,331 SPOOF: Sum-Product Optimization and Operator Fusion for Large-Scale Machine Learning 2017 CIDR
9 416 SystemML: Declarative Machine Learning on Spark 2016 VLDB
10 1,082 Hybrid Parallelization Strategies for Large-Scale Machine Learning in SystemML 2014 VLDB