The last decade has seen an exponential growth in the generation of a wide variety of data in an equally wide variety of settings (industrial, commercial, scientific, and personal). In order to successfully exploit this so-called Big-data wave (e.g., for competitive decision making), advanced analytic methods that turn raw data into meaningful and useful information are primordial and represent a major call to arms in Computer Science research. In particular, given the Velocity characteristic of Big Data, there is a growing need for so-called dynamic or reactive analytics methods that can support analytics in the presence of real-time updates.

This research project, graciously funded by the Wiener-Anspach foundation, aims to investigate fundamental and system challenges in supporting scalable analytics in the presence of real-time updates. The analytics problems envisioned cover wide areas of computer science and include database aggregate queries, probabilistic inference, matrix chain computation, and building statistical models. Our aim is to propose a uniform framework to dynamically solve this host of problem settings, focusing on contributing novel dynamic-algorithmic insights with attention to worst-case optimality, and the realisation of these algorithms in open-source software.

The project's duration is two years starting October 2016.

The Team

DFAQ is a joint research project between the Laboratory for Web and Information Technology at the Computer and Decision Engineering Department of the ULB and the Information Systems group at the Department of Computer Science of the University of Oxford. DFAQ is funded by the Wiener-Anspach foundation. Our development is also supported by EJ Technologies, in terms of donated licenses for their JProfiler Java Profiler.

Principal investigators:

Researchers :



  • Incremental Techniques for Large-Scale Dynamic Query Processing. CIKM 2018 Tutorial.
  • Incremental View Maintenance with Triple Lock Factorization Benefits. SIGMOD 2018.
  • Incremental View Maintenance with Triple Lock Factorization Benefits. At TorontoCon, Toronto, Canada in October 2017 and Computer Science Seminar, Université Libre de Bruxelles, Belgium in December 2017.
  • The Dynamic Yannakakis Algorithm. At SIGMOD 2017 Conference, Chicago, USA on May 18, 2017; Dresden Database Seminar, TU Dresden, Germany in April 2017; and Oxford Information Systems Seminar, Oxford, UK in November, 2017.
  • Incremental Evaluation of Updates in In-Memory Databases. Dutch-Belgian Database Day 2016, Mons, Belgium. October 28, 2016.