Our group focuses on the design and optimization of high-performance stream processing systems. We are dedicated to advancing the state of the art in various domains, including streaming databases, stream machine learning, stream data mining, and stream IoT. By leveraging cutting-edge techniques in concurrency control, adaptive scheduling, and data compression, we aim to address the scalability and efficiency challenges posed by real-time data streams. Our research contributions have been recognized in top-tier conferences and journals, reflecting our commitment to innovation and excellence in the field of stream data processing. If you like to join our team, please first answer a few questions in questionnaire.

[New] The PI, Dr. Zhang, will join the School of Computer Science and Technology at Huazhong University of Science and Technology (HUST) as a Professor in November 2024. We continue to welcome talented students to join our research team.

Streaming Database

Stream Processing Engine

• [ICDE’23] CompressStreamDB: Fine-Grained Adaptive Stream Processing without Decompression
• [TPDS’21] Fine-Grained Multi-Query Stream Processing on Integrated Architectures
• [ATC’20] FineStream: Fine-Grained Window-Based Stream Processing on CPU-GPU Integrated Architectures
• [SIGMOD Record’20] Hardware-Conscious Stream Processing: A Survey
• [SIGMOD’19] Briskstream: Scaling Data Stream Processing on Multicore Architectures
• [ICDE’17] Revisiting the Design of Data Stream Processing Systems on Multi-Core Processors
• [ICDE’17] Multi-Query Optimization for Complex Event Processing in SAP ESP

Stream Window Join

• [SIGMOD’25] Enabling Adaptive Sampling for Intra-Window Join: Simultaneously Optimizing Quantity and Quality
• [SIGMOD’24] PECJ: Stream Window Join on Disorder Data Streams with Proactive Error Compensation
• [ICDE’23] Scalable Online Interval Join on Modern Multicore Processors in OpenMLDB
• [SIGMOD’21] Parallelizing Intra-Window Join on Multicores: An Experimental Study

Stream Transactions

• [arXiv’24] TransNFV: Scalable State Management in NFV with Adaptive Transactional Concurrency Control
• [ICDE’24] MorphStream: Scalable Processing of Transactions over Streams
• [ICDE’24] Fast Parallel Recovery for Transactional Stream Processing on Multicores
• [VLDBJ’23] A Survey on Transactional Stream Processing
• [SIGMOD’23] MorphStream: Adaptive Scheduling for Scalable Transactional Stream Processing on Multicores
• [ICDE’20] Towards Concurrent Stateful Stream Processing on Multicore Processors

Stream Machine Learning/LLM, RAG

• [arXiv’24] StreamPrompt: Learnable Prompt-guided Data Selection for Efficient Stream Learning
• [arXiv’24] CANDY: A Benchmark for Continuous Approximate Nearest Neighbor Search with Dynamic Data Ingestion
• [arXiv’24] LibAMM: Empirical Insights into Approximate Computing for Accelerating Matrix Multiplication
• [arXiv’24] Online Continual Knowledge Learning for Language Models
• [EMNLP’23] SentiStream: A Co-Training Framework for Adaptive Online Sentiment Analysis in Evolving Data Streams

Stream Data Mining

• [ICDM’24] MOStream: A Modular and Self-Optimizing Data Stream Clustering Algorithm
• [SIGMOD’23] Data Stream Clustering: An In-depth Empirical Study
• [BigMM’19] TraV: an Interactive Trajectory Exploration System for Massive Data Sets

Stream IoT

• [TKDE’24] CStream: Parallel Data Stream Compression on Multicore Edge Devices
• [ICDE’23] Parallelizing Stream Compression for IoT Applications on Asymmetric Multicores
• [DEBS’23] A Hardware-Conscious Stateful Stream Compression Framework for IoT Applications (Vision)
• [VLIoT’20] NebulaStream: Complex Analytics Beyond the Cloud

click to see our other ancillary topics

Previous Next

At any time, we have a limited number of slots available. Unfortunately, we cannot respond to every message we receive. To help manage the process, we are using the questionnaire. We guarantee we will read the responses to this form but cannot guarantee a response. However, an impressive set of answers to this questionnaire is much more likely to result in a response than almost all other forms of contact.

If you are interested in pursuing graduate studies at NTU, please apply at the portal.