DBScholar

Back to papers

Production Machine Learning Pipelines: Empirical Analysis and Optimization Opportunities

Summary: Analyzes 3,000 production ML pipelines at Google via provenance graphs and 450k trainings to characterize lifespan, topology, and complexity. Introduces model graphlets, a data model for repeated components, and shows optimization opportunities—pruning wasted computation can cut costs by ~50% without delaying deployment. (summarized by gpt-5-nano on Feb 09 2026)

Paper ID
h2094ded95d203b3f
Venue
SIGMOD
Year
2021
Pagerank
9.0047829e-05
Overall Rank
2,116 | 85.82%
DOI
10.1145/3448016.3457566
PDF
Download (CC BY 4.0)

Incoming Non-self Citations Over Time

Authors

BibTeX Citation

@inproceedings{xin_sigmod21,
        title = {{Production Machine Learning Pipelines: Empirical Analysis and Optimization Opportunities}},
        author = {Xin, Doris and Miao, Hui and Parameswaran, Aditya and Polyzotis, Neoklis},
        series = {{SIGMOD} '21},
        booktitle = {Proceedings of the {ACM} {SIGMOD} International Conference on Management of Data},
        publisher = {Association for Computing Machinery},
        doi = {10.1145/3448016.3457566},
        url = {https://dl.acm.org/doi/10.1145/3448016.3457566},
        year = {2021}
}

Incoming Citations (Sorted by Pagerank)

Showing 17 of 17 citing papers.

Rank Citing Paper Year Venue Pagerank
4,949 TPCx-AI - An Industry Standard Benchmark for Artificial Intelligence and Machine Learning Systems 2023 VLDB 6.3271767e-05
6,438 CtxPipe: Context-aware Data Preparation Pipeline Construction for Machine Learning 2024 SIGMOD 5.7678369e-05
6,679 UPLIFT: Parallelization Strategies for Feature Transformations in Machine Learning Workloads 2022 VLDB 5.6956362e-05
7,558 Automating and Optimizing Data-Centric What-If Analyses on Native Machine Learning Pipelines 2023 SIGMOD 5.4788791e-05
8,407 AWARE: Workload-aware, Redundancy-exploiting Linear Algebra 2023 SIGMOD 5.3219903e-05
8,998 Pipemizer: An Optimizer for Analytics Data Pipelines 2022 VLDB 5.2273528e-05
9,099 Scheduling Data Processing Pipelines for Incremental Training on MLP-based Recommendation Models 2025 SIGMOD 5.208561e-05
9,146 Modyn: Data-Centric Machine Learning Pipeline Orchestration 2025 SIGMOD 5.2065925e-05
10,397 FlowPipe: LLM-Enhanced Conditional Generative Flow Networks for Data Preparation Pipeline Construction 2027 SIGMOD 4.9605343e-05
10,925 PipeLens: Identifying Interventions for Resolving Malfunctioning Data Science Pipelines 2026 VLDB 4.9605343e-05
10,968 stratum: A System Infrastructure for Massive Agent-Centric ML Workloads 2026 VLDB 4.9605343e-05
11,002 IMLane: Composable Framework for Efficient AI Function Execution in Database Engine 2026 VLDB 4.9605343e-05
11,031 SemPiper: Interactive Code Synthesis for Semantic Operators in Machine Learning Pipelines 2026 VLDB 4.9605343e-05
11,469 APEX-DAG: Library and Language independent Pipeline EXtraction 2025 VLDB 4.9605343e-05
11,643 Efficiently Mitigating the Impact of Data Drift on Machine Learning Pipelines 2024 VLDB 4.9605343e-05
11,781 Demystifying the QoS and QoE of Edge-hosted Video Streaming Applications in the Wild with SNESet 2023 SIGMOD 4.9605343e-05
11,804 Enabling Secure and Efficient Data Analytics Pipeline Evolution with Trusted Execution Environment 2023 VLDB 4.9605343e-05
Previous Page 1 / 1 Next

Outgoing Citations (Sorted by Pagerank)

Showing 14 of 14 cited papers.

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

Rank Cited Paper Year Venue Pagerank
105 The MADlib Analytics Library or MAD Skills, the SQL 2012 VLDB 0.00033605348
512 Goods: Organizing Google's Datasets 2016 SIGMOD 0.00017036842
1,151 Data Management Challenges in Production Machine Learning 2017 SIGMOD 0.000117662
1,562 Titian: Data Provenance Support in Spark 2016 VLDB 0.00010192248
1,567 HELIX: Holistic Optimization for Accelerating Iterative Machine Learning 2019 VLDB 0.00010179013
1,776 Putting Lipstick on Pig: Enabling Database-style Workflow Provenance 2012 VLDB 9.6661217e-05
1,878 Ground: A Data Context Service 2017 CIDR 9.431234e-05
1,938 Elastic Machine Learning Algorithms in Amazon SageMaker 2020 SIGMOD 9.3316919e-05
2,199 Extending Relational Query Processing with ML Inference 2020 CIDR 8.8664379e-05
2,448 noWorkflow: a Tool for Collecting, Analyzing, and Managing Provenance from Python Scripts 2017 VLDB 8.4366124e-05
4,174 The Relational Data Borg is Learning 2020 VLDB 6.750725e-05
4,802 Improving Reproducibility of Data Science Pipelines through Transparent Provenance Capture 2020 VLDB 6.3924551e-05
5,748 An Optimal Labeling Scheme for Workflow Provenance Using Skeleton Labels 2010 SIGMOD 5.9914851e-05
6,213 Incremental View Maintenance For Collection Programming 2016 PODS 5.8312546e-05
Previous Page 1 / 1 Next

Semantically Similar Papers

# Overall Rank Paper Year Venue
1 2,062 tf.data: A Machine Learning Data Processing Framework 2021 VLDB
2 3,598 Where Is My Training Bottleneck? Hidden Trade-Offs in Deep Learning Preprocessing Pipelines 2022 SIGMOD
3 12,175 Leveraging Organizational Resources to Adapt Models to New Data Modalities 2020 VLDB
4 5,808 Optimizing Machine Learning Workloads in Collaborative Environments 2020 SIGMOD
5 11,875 Data Management Opportunities for Foundation Models 2022 CIDR
6 11,871 Towards Observability for Machine Learning Pipelines 2022 CIDR
7 7,235 Capturing and Querying Fine-grained Provenance of Preprocessing Pipelines in Data Science 2021 VLDB
8 1,151 Data Management Challenges in Production Machine Learning 2017 SIGMOD
9 9,454 Towards Observability for Production Machine Learning Pipelines 2022 VLDB
10 6,234 Materialization and Reuse Optimizations for Production Data Science Pipelines 2022 SIGMOD