Scheduling Data Processing Pipelines for Incremental Training on MLP-based Recommendation Models

Summary: Intra-pipeline prefetching hides feature processing during embedding lookup in incremental MLP training on CPU-limited clusters. Inter-pipeline scheduling overlaps critical comms with idle non-critical tasks, yielding 1.36x speedup via RECS on TensorFlow. (summarized by gpt-5-nano on Feb 09 2026)

Paper ID: 7179
Venue: SIGMOD
Year: 2025
Pagerank: 5.3483178e-05
Overall Rank: 8,907 | 38.90%
DOI: 10.1145/3722212.3724454

Incoming Non-self Citations Over Time

Authors

1. Zihao Chen (East China Normal University)
2. Chenyang Zhang (East China Normal University)
3. Chen Xu (East China Normal University)
4. Zhao Zhang (East China Normal University)
5. Jiaqiang Wang (Tencent)
6. Weining Qian (East China Normal University)
7. Aoying Zhou (East China Normal University)

BibTeX Citation

@inproceedings{chen_sigmod25,
        title = {{Scheduling Data Processing Pipelines for Incremental Training on MLP-based Recommendation Models}},
        author = {Chen, Zihao and Zhang, Chenyang and Xu, Chen and Zhang, Zhao and Wang, Jiaqiang and Qian, Weining and Zhou, Aoying},
        series = {{SIGMOD} '25},
        booktitle = {Proceedings of the {ACM} {SIGMOD} International Conference on Management of Data},
        publisher = {Association for Computing Machinery},
        doi = {10.1145/3722212.3724454},
        url = {https://dl.acm.org/doi/10.1145/3722212.3724454},
        year = {2025}
}

Incoming Citations (Sorted by Pagerank)

Showing 1 of 1 citing papers.

Rank	Citing Paper	Year	Venue	Pagerank
10,472	Mixtera: A Data Plane for Foundation Model Training	2026	SIGMOD	5.093636e-05

Outgoing Citations (Sorted by Pagerank)

Showing 13 of 13 cited papers.

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

Rank	Cited Paper	Year	Venue	Pagerank
1,150	DimmWitted: A Study of Main-Memory Statistical Analytics	2014	VLDB	0.00011943462
1,446	Analyzing and Mitigating Data Stalls in DNN Training	2021	VLDB	0.0001076818
2,018	tf.data: A Machine Learning Data Processing Framework	2021	VLDB	9.3001937e-05
2,485	HET: Scaling out Huge Embedding Model Training via Cache-enabled Distributed Framework	2022	VLDB	8.5145736e-05
2,657	Production Machine Learning Pipelines: Empirical Analysis and Optimization Opportunities	2021	SIGMOD	8.2887895e-05
2,688	Accelerating Recommendation System Training by Leveraging Popular Choices	2022	VLDB	8.2564305e-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
4,912	HET-GMP: A Graph-based System Approach to Scaling Large Embedding Model Training	2022	SIGMOD	6.4481656e-05
5,282	GoldMiner: Elastic Scaling of Training Data Pre-Processing Pipelines for Deep Learning	2023	SIGMOD	6.2840712e-05
6,309	Materialization and Reuse Optimizations for Production Data Science Pipelines	2022	SIGMOD	5.9189554e-05
7,962	MultiBiSage: A Web-Scale Recommendation System Using Multiple Bipartite Graphs at Pinterest	2023	VLDB	5.5180565e-05
9,164	FEC: Efficient Deep Recommendation Model Training with Flexible Embedding Communication	2023	SIGMOD	5.3101785e-05

Semantically Similar Papers

#	Overall Rank	Paper	Year	Venue
1	3,764	FastFlow: Accelerating Deep Learning Model Training with Smart Offloading of Input Data Pipeline	2023	VLDB
2	4,529	TencentRec: Real-time Stream Recommendation in Practice	2015	SIGMOD
3	2,929	StreamRec: A Real-Time Recommender System	2011	SIGMOD
4	6,309	Materialization and Reuse Optimizations for Production Data Science Pipelines	2022	SIGMOD
5	2,018	tf.data: A Machine Learning Data Processing Framework	2021	VLDB
6	3,541	Where Is My Training Bottleneck? Hidden Trade-Offs in Deep Learning Preprocessing Pipelines	2022	SIGMOD
7	6,954	DLRover-RM: Resource Optimization for Deep Recommendation Models Training in the Cloud	2024	VLDB
8	9,164	FEC: Efficient Deep Recommendation Model Training with Flexible Embedding Communication	2023	SIGMOD
9	10,804	IncrCP: Decomposing and Orchestrating Incremental Checkpoints for Effective Recommendation Model Training	2025	VLDB
10	2,688	Accelerating Recommendation System Training by Leveraging Popular Choices	2022	VLDB