PS2: Parameter Server on Spark

Summary: PS2 deploys a parameter server on Spark, preserving Spark for data processing while offloading models, without hacking Spark core. DCV enables locality-aware, element-wise multi-vector ops, delivering up to 55.6x Spark MLlib and 3.7x Petuum. (summarized by gpt-5-nano on Feb 09 2026)

Paper ID: 5712
Venue: SIGMOD
Year: 2019
Pagerank: 6.2586047e-05
Overall Rank: 5,345 | 63.33%
DOI: 10.1145/3299869.3314038

Incoming Non-self Citations Over Time

Authors

1. Zhipeng Zhang (Peking University; Tencent)
2. Bin Cui (Peking University)
3. Yingxia Shao (Beijing Institute of Technology)
4. Lele Yu (Tencent)
5. Jiawei Jiang (Tencent)
6. Xupeng Miao (Peking University; Tencent)

BibTeX Citation

@inproceedings{zhang_sigmod19,
        title = {{PS2: Parameter Server on Spark}},
        author = {Zhang, Zhipeng and Cui, Bin and Shao, Yingxia and Yu, Lele and Jiang, Jiawei and Miao, Xupeng},
        series = {{SIGMOD} '19},
        booktitle = {Proceedings of the {ACM} {SIGMOD} International Conference on Management of Data},
        publisher = {Association for Computing Machinery},
        doi = {10.1145/3299869.3314038},
        url = {https://dl.acm.org/doi/10.1145/3299869.3314038},
        year = {2019}
}

Incoming Citations (Sorted by Pagerank)

Showing 8 of 8 citing papers.

Rank	Citing Paper	Year	Venue	Pagerank
2,485	HET: Scaling out Huge Embedding Model Training via Cache-enabled Distributed Framework	2022	VLDB	8.5145736e-05
4,067	Distributed Deep Learning on Data Systems: A Comparative Analysis of Approaches	2021	VLDB	6.9293511e-05
4,956	Heterogeneity-Aware Distributed Machine Learning Training via Partial Reduce	2021	SIGMOD	6.4290135e-05
5,589	An Experimental Evaluation of Large Scale GBDT Systems	2019	VLDB	6.1559057e-05
5,876	BAGUA: Scaling up Distributed Learning with System Relaxations	2022	VLDB	6.0557672e-05
6,022	NuPS: A Parameter Server for Machine Learning with Non-Uniform Parameter Access	2022	SIGMOD	6.0046123e-05
6,519	Dynamic Parameter Allocation in Parameter Servers	2020	VLDB	5.8536062e-05
8,334	Just Move It! Dynamic Parameter Allocation in Action	2021	VLDB	5.4518674e-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
24	Spark SQL: Relational Data Processing in Spark	2015	SIGMOD	0.00054865648
345	Column-Stores vs. Row-Stores: How Different Are They Really?	2008	SIGMOD	0.00020656723
415	SystemML: Declarative Machine Learning on Spark	2016	VLDB	0.0001888524
2,162	Heterogeneity-aware Distributed Parameter Servers	2017	SIGMOD	9.0581831e-05
9,672	DimBoost: Boosting Gradient Boosting Decision Tree to Higher Dimensions	2018	SIGMOD	5.2386185e-05
11,999	LDA*: A Robust and Large-scale Topic Modeling System	2017	VLDB	5.093636e-05

Semantically Similar Papers

#	Overall Rank	Paper	Year	Venue
1	6,850	Bridging the Gap Between HPC and Big Data Frameworks	2017	VLDB
2	8,535	sPCA: Scalable Principal Component Analysis for Big Data on Distributed Platforms	2015	SIGMOD
3	11,392	ST4ML: Machine Learning Oriented Spatio-Temporal Data Processing at Scale	2023	SIGMOD
4	7,190	PetPS: Supporting Huge Embedding Models with Persistent Memory	2023	VLDB
5	1,079	Hybrid Parallelization Strategies for Large-Scale Machine Learning in SystemML	2014	VLDB
6	2,162	Heterogeneity-aware Distributed Parameter Servers	2017	SIGMOD
7	8,615	A Spark Optimizer for Adaptive, Fine-Grained Parameter Tuning	2024	VLDB
8	415	SystemML: Declarative Machine Learning on Spark	2016	VLDB
9	6,344	Towards General and Efficient Online Tuning for Spark	2023	VLDB
10	2,693	FlexPS: Flexible Parallelism Control in Parameter Server Architecture	2018	VLDB