Site moved to


  • Computational Journalism: leverage computing to help preserving public interest journalism.
  • PQ: perturbation analysis of database queries.
  • Cumulon: simplifying the development and deployment of statistical analysis programs in the cloud with automatic optimization and provisioning.
  • Flex: a platform for experiment-driven system management.
  • Proteusa practical and rigorous toolkit for private data analysis
  • Ques: querying and controlling systems. DIADS is subproject looking specifically at the integrated management of database and storage systems.
  • Starfish: a self-tuning system for big data analytics.


  • RIOT: transparently bringing scalability and I/O-efficiency to statistical computing with R; i.e., no need to rewrite your R code! 2009-2014.
  • DDDAS: dynamic data-driven environmental sensor network in Duke Forest (collaboration with Duke School of the Environment). 2006-2012.
  • ProSem: Internet-scale publish/subscribe unifying data processing and dissemination. 2007-2011.
  • ERS: tracking and exploring lineage in experimental and computational workflows for biomedical research (collaboration with Duke Center for Computational Immunology). 2005-2011.
  • DDM: techniques and applications of maintaining various forms of derived data (e.g., caches, replicas, indexes, materialized views, synopses). 2001-2008.