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)
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 |
|---|---|---|---|---|
| 2,519 | HET: Scaling out Huge Embedding Model Training via Cache-enabled Distributed Framework | 2022 | VLDB | 8.3521095e-05 |
| 4,095 | Distributed Deep Learning on Data Systems: A Comparative Analysis of Approaches | 2021 | VLDB | 6.8095767e-05 |
| 4,963 | Heterogeneity-Aware Distributed Machine Learning Training via Partial Reduce | 2021 | SIGMOD | 6.3384091e-05 |
| 5,633 | NuPS: A Parameter Server for Machine Learning with Non-Uniform Parameter Access | 2022 | SIGMOD | 6.0578661e-05 |
| 5,723 | An Experimental Evaluation of Large Scale GBDT Systems | 2019 | VLDB | 6.0178515e-05 |
| 5,997 | BAGUA: Scaling up Distributed Learning with System Relaxations | 2022 | VLDB | 5.9198921e-05 |
| 6,640 | Dynamic Parameter Allocation in Parameter Servers | 2020 | VLDB | 5.7251952e-05 |
| 8,488 | Just Move It! Dynamic Parameter Allocation in Action | 2021 | VLDB | 5.3317701e-05 |
Previous
Page 1 / 1
Next
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 |
|---|---|---|---|---|
| 23 | Spark SQL: Relational Data Processing in Spark | 2015 | SIGMOD | 0.00055406774 |
| 331 | Column-Stores vs. Row-Stores: How Different Are They Really? | 2008 | SIGMOD | 0.0002077683 |
| 415 | SystemML: Declarative Machine Learning on Spark | 2016 | VLDB | 0.0001865959 |
| 2,184 | Heterogeneity-aware Distributed Parameter Servers | 2017 | SIGMOD | 8.8958335e-05 |
| 9,846 | DimBoost: Boosting Gradient Boosting Decision Tree to Higher Dimensions | 2018 | SIGMOD | 5.121318e-05 |
| 12,296 | LDA*: A Robust and Large-scale Topic Modeling System | 2017 | VLDB | 4.9793485e-05 |
Previous
Page 1 / 1
Next
Semantically Similar Papers
| # | Overall Rank | Paper | Year | Venue |
|---|---|---|---|---|
| 1 | 6,996 | Bridging the Gap Between HPC and Big Data Frameworks | 2017 | VLDB |
| 2 | 8,704 | sPCA: Scalable Principal Component Analysis for Big Data on Distributed Platforms | 2015 | SIGMOD |
| 3 | 11,707 | ST4ML: Machine Learning Oriented Spatio-Temporal Data Processing at Scale | 2023 | SIGMOD |
| 4 | 7,333 | PetPS: Supporting Huge Embedding Models with Persistent Memory | 2023 | VLDB |
| 5 | 1,081 | Hybrid Parallelization Strategies for Large-Scale Machine Learning in SystemML | 2014 | VLDB |
| 6 | 2,184 | Heterogeneity-aware Distributed Parameter Servers | 2017 | SIGMOD |
| 7 | 8,004 | A Spark Optimizer for Adaptive, Fine-Grained Parameter Tuning | 2024 | VLDB |
| 8 | 415 | SystemML: Declarative Machine Learning on Spark | 2016 | VLDB |
| 9 | 6,450 | Towards General and Efficient Online Tuning for Spark | 2023 | VLDB |
| 10 | 2,725 | FlexPS: Flexible Parallelism Control in Parameter Server Architecture | 2018 | VLDB |