Home Personal Activities Students & Postdocs Presentations Publications Vita Teaching Research Awards

Research

My research is on data management. These days I would characterize it as data engineering aspects of data science. The main focus of my research follows two threads: (1) application of database technology to non-traditional data types, and (2) distributed & parallel data management. These two threads usually converge.

Investigating how database technology can be applied to data types that are more complex than business data processing (for which relational systems are the perfect fit) has always been one of my interests. Frank Tompa characterizes this type of work as Data Management(X) where X is the data type of interest -- we also have a graduate course with exactly this focus (CS 741). At different times in my career, X has been equal to one or more of the following: "object" data (in the sense of object databases), multimedia data, temporal data, spatial data, XML, stream data. Currently, my focus is on graph data and RDF data.

Graph data management

Graphs have always been important data types for database researchers. With the recent growth of social networks, Wikipedia, Linked Data, RDF, and other networks, the interest in managing very large graphs have again gained momentum. I have a number of projects in this space.

Relevant Projects

Publications

  1. A. Sheshbolouki and M. T. Özsu, sGrow: Explaining the Scale-Invariant Strength Assortativity of Streaming Butterflies, ACM Transactions on the Web, 2022. Accepted for publication.
  2. L. Zheng, L. Zou and M. T. Özsu. SGSI – A Scalable GPU-friendly Subgraph Isomorphism Algorithm. IEEE Trans. Knowledge and Data Eng., 2022. Accepted for publication
  3. A. Pacaci, A. Bonifati and M.T. Özsu, Evaluating Complex Queries on Streaming Graphs, In Proc. 28th IEEE Int. Conf. on Data Eng., pages 272-285, 2022.
  4. K. Ammar, S. Sadhu, S. Salihoglu and M. T. Özsu. Optimizing Differentially-Maintained Recursive Queries on Dynamic Graphs. Proc. VLDB Endowment, 15(11): 3186-3198, 2022.
  5. A. Sheshbolouki and M. T. Özsu. sGrapp: Butterfly Approximation in Streaming Graphs, ACM Transactions on Knowledge Discovery From Data, 16(4): Article 76, 2022.
  6. A. Pacaci, A. Bonifati and M. T. Özsu.Regular Path Query Evaluation on Streaming Graphs, In Proc. ACM SIGMOD International Conference on Management of Data, pages 1415-1430, 2020.
  7. L. Zeng, L. Zou, M. T.Özsu, L. Hu, and F. Zhang. GSI: GPU-friendly Subgraph Isomorphism. In Proc. 36th International Conference on Data Engineering, pages 1249-1260, 2020.
  8. A. Sahu, A. Mhedhbi, S. Salihoglu, J. Lin, M. T. Özsu.The Ubiquity of Large Graphs and Surprising Challenges of Graph Processing: Extended Survey, VLDB Journal, 29: 595–618, 2020.
  9. A. Pacaci and M. T. Özsu.Analysis of Streaming Algorithms for Graph Partitioning, In Proc. ACM SIGMOD International Conference on Management of Data, pages 1375–1392, 2019.
  10. X. Li and M. T. Özsu.Correlation Constraint Shortest Path over Large Multi-Correlation Graphs, Proc. VLDB Endowment, 12(5): 488-501, 2019.
  11. A. Sahu, A. Mhedhbi, S. Salihoglu, J. Lin, M. T. Özsu. The Ubiquity of Large Graphs and Surprising Challenges of Graph Processing, Proc. VLDB Endowment, 11(4): 420-431, 2018.

Other publications on this topic can be found here.

RDF data management

Resource Description Framework (RDF) has been proposed for modeling Web objects as part of developing the semantic web. It has also gained attention as a way to accomplish web data integration. For example, the Linking Open Data (LOD) cloud is a distributed RDF knowledge base created over hundreds of autonomous datasets. Currently, the LOD cloud contains more than 25 billion triples, and its size is doubling every year. As the volume of RDF data has increased, interesting data management issues have arisen. We study the processing of SPARQL queries over RDF data using a graph-theoretic approach: we represent both the RDF data and the SPARQL queries a graphs and conver the query evaluation problem toone of subgraph matching. My work in this area covers a number of topics:

Relevant Projects

     

Publications

  1. P. Peng, M. T. Özsu, L. Zou, C. Yan, C.Liu. MPC: Minimum Property-Cut RDF Graph Partitioning. In Proc. 28th IEEE Int. Conf. on Data Eng., pages 192-204, 2022.
  2. P. Peng, Q. Ge, L. Zou, M. T. Özsu, Z. Xu, and D. Zhao. Optimizing Multi-Query Evaluation in Federated RDF Systems, IEEE Trans. Knowledge and Data Eng., 33(4):1692–1707, 2021.
  3. G. Aluç, M. T. Özsu, and K. Daudjee. Clustering RDF Databases Using Tunable-LSH, VLDB Journal, 28(2): 173-195, 2019.
  4. L. Gao, L. Golab, M. T. Özsu, G. Aluç. Stream WatDiv: A Streaming RDF Benchmark, In Proc. International Workshop on Semantic Big Data, pages 1-6, 2018.
  5. O. Hartig and M. T. Özsu. Walking without a Map: Ranking-Based Traversal for Querying Linked Data, In Proc. 15th International Semantic Web Conference, pages 305–324, 2016.
  6. P. Peng, L. Zou, M. T. Özsu, L. Chen, D. Zhao.Processing SPARQL Queries Over Distributed RDF Graphs, VLDB Journal, 25(2):243–268, 2016.
  7. G. Aluç, M. T. Özsu, K. Daudjee, and O. Hartig.Executing queries over schemaless RDF databases, In Proc. 31st Int. Conf. on Data Engineering, pages 807 - 818, 2015.
  8. L. Zou, M. T. Özsu, L. Chen, X. Sheng, R. Huang, and D. Zhao.gStore: A Graph-based SPARQL Query Engine, VLDB Journal, 23(4): 565-590, 2014.
  9. G. Aluç, O. Hartig, M. T. Özsu, and K. Daudjee. Diversified stress testing of RDF data management systems, In Proc. 13th Int. Semantic Web Conference, Part I, pages 197–212, 2014.
  10. G. Aluç, M. T. Özsu, and K. Daudjee. Workload matters: Why RDF databases need a new design, Proc. VLDB Endowment, 7(10):837–840, 2014.
  11. G. Aluç, M. T. Özsu, K. Daudjee, and O. Hartig. chameleon-db: a workload-aware robust RDF data management system,Technical Report CS-2013-10, University of Waterloo, 2013.

Other publications on this topic can be found here.

Scale-out data management

Work in this area typically overlaps with the previous two topics as I investigate how graph processing and RDF data management scale-out over distributed and parallel systems.

Publications

  1. R. Wang, J. Wang, S. Idreos, M.T. Özsu, W. G. Aref. The Case for Distributed Shared-Memory Databases with RDMA-Enabled Memory Disaggregation, Proc. VLDB Endowment, 16(1): 15-22, 2022.
  2. D. Yan, G. Guo, J. Khalil, M. T. Özsu, Wei-Shinn Ku, and John C.S. Lui. G-thinker: A general distributed framework for finding qualified subgraphs in a big graph with load balancing. VLDB Journal, 31: 287–320, 2022
  3. P. Valduriez, R. Jimenez-Peris, and M. T. Özsu. Distributed database systems: The case for NewSQL. In Abdelkader Hameurlain and A. Min Tjoa, editors, Transactions on Large-Scale Data- and Knowledge-Centered Systems, pages 1–15. Springer, Berlin, Heidelberg, 2021.
  4. A. Pacaci and M.T. Özsu. Experimental Analysis of Streaming Algorithms for Graph Partitioning, In Proc. ACM SIGMOD Int. Conf. Management of Data, pages 1375-1392, 2019.
  5. K. Ammar and M. T. Özsu. Experimental Analysis of Distributed Graph Systems, Proc. VLDB Endowment, 11(10): 1151-1164, 2018.
  6. S. Salihoglu and M. T. Özsu. Response to `Scale Up or Scale Out for Graph Processing?', IEEE Internet Comput., 22(5):18–24, 2018.
  7. D. Yan, J. Cheng, M. T. Özsu, F. Yang, Y. Lu, J. C.S. Liu, Q. Zhang, and W. Ng. A General-Purpose Query-Centric Framework for Querying Big Graphs. Proc. VLDB Endowment, 9(7):564 – 575, 2016.

Other publications on this topic can be found here.

Last update: