AllianceDB is an open-source benchmark suite for evaluating stream join algorithms on multicore hardware. It collects reproducibility artifacts for several related join studies; it is a DataSys benchmark suite, not a single production database runtime.
The root scripts coordinate multiple historical experiment branches and hardware-sensitive measurements. Start with the artifact documentation and review its CPU, NUMA, and privilege requirements before running experiments.
The repository is licensed under the MIT License. Bundled third-party source files retain their upstream notices.
If you use AllianceDB in your work, please cite our work.
- [SIGMOD 2021] Shuhao Zhang, Yancan Mao, Jiong He, Philipp M. Grulich, Steffen Zeuch, Bingsheng He, Richard T. B. Ma, and Volker Markl. Parallelizing Intra-Window Join on Multicores: An Experimental Study. Artifact release.
- [ICDE 2023] Hao Zhang, Xianzhi Zeng, Shuhao Zhang, Xinyi Liu, Mian Lu, and Zhao Zheng. Scalable Online Interval Join on Modern Multicore Processors in OpenMLDB.
- [SIGMOD 2024] Xianzhi Zeng, Shuhao Zhang, Hongbin Zhong, Hao Zhang, Mian Lu, Zhao Zheng, and Yuqiang Chen. PECJ: Stream Window Join on Disorder Data Streams with Proactive Error Compensation.
- [PACMMOD 2024 / SIGMOD 2025] Xilin Tang, Feng Zhang, Shuhao Zhang, Yani Liu, Bingsheng He, and Xiaoyong Du. Enabling Adaptive Sampling for Intra-Window Join: Simultaneously Optimizing Quantity and Quality.