tf.data: A Machine Learning Data Processing Framework

Summary: tf.data is a compositional framework and runtime for high-throughput ML input pipelines, hiding parallelism, caching, overlap, and tuning. Fleet-scale analysis shows diverse preprocessing consumes substantial training resources, motivating cross-job sharing and storage-level projection. (summarized by gpt-5.6-luna on Jul 24 2026)

Paper ID: 12691
Venue: VLDB
Year: 2021
Pagerank: 9.3001937e-05
Overall Rank: 2,018 | 86.16%
DOI: 10.14778/3476311.3476374

Incoming Non-self Citations Over Time

Authors

1. Derek G. Murray (Lacework)
2. Jiří Šimša (Google)
3. Ana Klimovic (ETH Zurich)
4. Ihor Indyk (Google)

BibTeX Citation

@article{murray_vldb21,
        title = {{tf.data: A Machine Learning Data Processing Framework}},
        author = {Murray, Derek G. and Šimša, Jiří and Klimovic, Ana and Indyk, Ihor},
        journal = {PVLDB},
        series = {{VLDB} '21},
        volume = {14},
        number = {12},
        pages = {2945--2958},
        doi = {10.14778/3476311.3476374},
        url = {https://doi.org/10.14778/3476311.3476374},
        year = {2021}
}

Incoming Citations (Sorted by Pagerank)

Showing 16 of 16 citing papers.

Rank	Citing Paper	Year	Venue	Pagerank
2,865	End-to-end Optimization of Machine Learning Prediction Queries	2022	SIGMOD	8.0180243e-05
3,541	Where Is My Training Bottleneck? Hidden Trade-Offs in Deep Learning Preprocessing Pipelines	2022	SIGMOD	7.3280673e-05
3,764	FastFlow: Accelerating Deep Learning Model Training with Smart Offloading of Input Data Pipeline	2023	VLDB	7.1446931e-05
5,282	GoldMiner: Elastic Scaling of Training Data Pre-Processing Pipelines for Deep Learning	2023	SIGMOD	6.2840712e-05
6,190	Progressive Compressed Records: Taking a Byte out of Deep Learning Data	2021	VLDB	5.9476635e-05
6,538	UPLIFT: Parallelization Strategies for Feature Transformations in Machine Learning Workloads	2022	VLDB	5.8477764e-05
7,285	Bullion: A Column Store for Machine Learning	2025	CIDR	5.6559604e-05
8,321	FusionFlow: Accelerating Data Preprocessing for Machine Learning with CPU-GPU Cooperation	2024	VLDB	5.4544352e-05
8,905	TensorSocket: Shared Data Loading for Deep Learning Training	2026	SIGMOD	5.3483178e-05
8,907	Scheduling Data Processing Pipelines for Incremental Training on MLP-based Recommendation Models	2025	SIGMOD	5.3483178e-05
8,949	Modyn: Data-Centric Machine Learning Pipeline Orchestration	2025	SIGMOD	5.3462965e-05
9,880	MEMO: Fine-grained Tensor Management For Ultra-long Context LLM Training	2025	SIGMOD	5.2040783e-05
10,472	Mixtera: A Data Plane for Foundation Model Training	2026	SIGMOD	5.093636e-05
10,509	FlatStor: An Efficient Embedded-Index Based Columnar Data Layout for Multimodal Data Workloads	2026	VLDB	5.093636e-05
10,844	GPEmu: A GPU Emulator for Faster and Cheaper Prototyping and Evaluation of Deep Learning System Research	2025	VLDB	5.093636e-05
10,999	cedar: Optimized and Unified Machine Learning Input Data Pipelines	2025	VLDB	5.093636e-05

Outgoing Citations (Sorted by Pagerank)

Showing 6 of 6 cited papers.

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

Rank	Cited Paper	Year	Venue	Pagerank
106	The MADlib Analytics Library or MAD Skills, the SQL	2012	VLDB	0.00033539462
292	LINQ: Reconciling Objects, Relations and XML in the .NET Framework	2006	SIGMOD	0.00022259549
1,446	Analyzing and Mitigating Data Stalls in DNN Training	2021	VLDB	0.0001076818
2,293	Extending Relational Query Processing with ML Inference	2020	CIDR	8.7949378e-05
3,146	HELIX: Accelerating Human-in-the-loop Machine Learning	2018	VLDB	7.7091591e-05
3,253	Jointly Optimizing Preprocessing and Inference for DNN-based Visual Analytics	2021	VLDB	7.5936939e-05

Semantically Similar Papers

#	Overall Rank	Paper	Year	Venue
1	8,321	FusionFlow: Accelerating Data Preprocessing for Machine Learning with CPU-GPU Cooperation	2024	VLDB
2	4,592	Data Platform for Machine Learning	2019	SIGMOD
3	5,282	GoldMiner: Elastic Scaling of Training Data Pre-Processing Pipelines for Deep Learning	2023	SIGMOD
4	6,538	UPLIFT: Parallelization Strategies for Feature Transformations in Machine Learning Workloads	2022	VLDB
5	2,196	Spinning Fast Iterative Data Flows	2012	VLDB
6	2,657	Production Machine Learning Pipelines: Empirical Analysis and Optimization Opportunities	2021	SIGMOD
7	3,541	Where Is My Training Bottleneck? Hidden Trade-Offs in Deep Learning Preprocessing Pipelines	2022	SIGMOD
8	10,999	cedar: Optimized and Unified Machine Learning Input Data Pipelines	2025	VLDB
9	3,630	TensorFlow Data Validation: Data Analysis and Validation in Continuous ML Pipelines	2020	SIGMOD
10	3,764	FastFlow: Accelerating Deep Learning Model Training with Smart Offloading of Input Data Pipeline	2023	VLDB