Model-Free Control for Distributed Stream Data Processing using Deep Reinforcement Learning

Summary: Introduces the first model-free deep-RL controller for distributed stream scheduling, learning workload placement from sparse runtime statistics without queueing models. Implemented in Storm, it jointly optimizes placement and communication effects, cutting tuple latency 33.5% over default scheduling. (summarized by gpt-5.6-luna on Jul 24 2026)

Paper ID: 11965
Venue: VLDB
Year: 2018
Pagerank: 5.6766532e-05
Overall Rank: 7,194 | 50.65%
DOI: 10.14778/3184470.3184474

Incoming Non-self Citations Over Time

Authors

1. Teng Li (Syracuse University)
2. Zhiyuan Xu (Syracuse University)
3. Jian Tang (Syracuse University)
4. Yanzhi Wang (Syracuse University)

BibTeX Citation

@article{li_vldb18,
        title = {{Model-Free Control for Distributed Stream Data Processing using Deep Reinforcement Learning}},
        author = {Li, Teng and Xu, Zhiyuan and Tang, Jian and Wang, Yanzhi},
        journal = {PVLDB},
        series = {{VLDB} '18},
        volume = {11},
        number = {6},
        pages = {705--718},
        doi = {10.14778/3184470.3184474},
        url = {https://doi.org/10.14778/3184470.3184474},
        year = {2018}
}

Incoming Citations (Sorted by Pagerank)

Showing 4 of 4 citing papers.

Rank	Citing Paper	Year	Venue	Pagerank
2,499	Learning a Partitioning Advisor for Cloud Databases	2020	SIGMOD	8.4993549e-05
4,468	One Model to Rule them All: Towards Zero-Shot Learning for Databases	2022	CIDR	6.6819041e-05
5,388	Fine-Grained Modeling and Optimization for Intelligent Resource Management in Big Data Processing	2022	VLDB	6.2362811e-05
11,065	Learned Cost Models for Query Optimization: From Batch to Streaming Systems	2025	VLDB	5.093636e-05

Outgoing Citations (Sorted by Pagerank)

Showing 2 of 2 cited papers.

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

Rank	Cited Paper	Year	Venue	Pagerank
224	MillWheel: Fault-Tolerant Stream Processing at Internet Scale	2013	VLDB	0.00024130894
2,357	Muppet: MapReduce-Style Processing of Fast Data	2012	VLDB	8.7013171e-05

Semantically Similar Papers

#	Overall Rank	Paper	Year	Venue
1	7,912	Data Stream Clustering: An In-depth Empirical Study	2023	SIGMOD
2	4,987	Releasing Cloud Databases from the Chains of Performance Prediction Models	2017	CIDR
3	8,454	D3-GNN: Dynamic Distributed Dataflow for Streaming Graph Neural Networks	2024	VLDB
4	2,499	Learning a Partitioning Advisor for Cloud Databases	2020	SIGMOD
5	9,128	Dalton: Learned Partitioning for Distributed Data Streams	2023	VLDB
6	13,375	Reimagining Deep Learning Systems Through the Lens of Data Systems	2024	VLDB
7	8,663	Scalable Delivery of Stream Query Result	2009	VLDB
8	334	An End-to-End Automatic Cloud Database Tuning System Using Deep Reinforcement Learning	2019	SIGMOD
9	10,547	Scarf: Self-Adaptive Tuning via Multi-Objective Reinforcement Learning for Apache Flink	2026	VLDB
10	4,067	Distributed Deep Learning on Data Systems: A Comparative Analysis of Approaches	2021	VLDB