Document Databases

Document-oriented databases are becoming more popular due to the prevalence of semi-structured data. The document model allows entities to be represented in a schema-less manner using a hierarchy of properties. Document DBMSs support rich relational and hierarchical queries over such documents.

We are working on several projects for adapating techniques from both relational DBMSs and other areas of software systems for document DBMSs. This includes automatic database design for schemaless applications, optimizations for replicated environments, and improving the efficiency of document-oriented storage managers.



This work is supported (in part) by an Amazon AWS Research Grant and MongoDB Inc.


  • L. Xu, A. Pavlo, S. Sengupta, J. Li, and G. R. Ganger, "Reducing Replication Bandwidth for Distributed Document Databases," in Proceedings of the Sixth ACM Symposium on Cloud Computing, 2015, pp. 222-235. [PDF] [BIBTEX]
      author = {Xu, Lianghong and Pavlo, Andrew and Sengupta, Sudipta and Li, Jin and Ganger, Gregory R.},
      title = {Reducing Replication Bandwidth for Distributed Document Databases},
      booktitle = {Proceedings of the Sixth ACM Symposium on Cloud Computing},
      series = {SoCC '15},
      year = {2015},
      pages = {222--235},
      numpages = {14},
      url = {},
Visit Project Homepage