Research Outline

I have mainly interest in issues linked to high-performance computing (HPC) and distributed applications (Big Data or Cloud Computing for example). I focus in particular on scheduling aspects, more precisely on static scheduling (computed before execution) based on theoretical models. During my (short) career I deal with several specific problems.


Work in progress (If you can read it, French version is more complete).


Among the new issues that appears with the increasing importance of high-performance computing and Big Data applications, the reduction of data transfers (communication between machines) is crucial as it can implies delay and increasing of energy consumption. Replication, of tasks or files, has an increasing impact on the amount of communication, however it is an useful tool to improve parallelism of computations and fault tolerance. During my PhD, with my supervisors Olivier Beaumont and Lionel Eyraud-Dubois, we were interested to input files replication and its impact on communication with two practical problems. In first one we consider parallel matrix product and try to decrease the number of replicated data created during the computation. In the second case, we were interested in the scheduling of a set of independent tasks that depend on input files that are, before the execution, already replicated and placed on different machines of the clusters. In such case, our goal was to use in best possible way this replication in order to have the best possible completion time or to avoid additional duplication.


Adapted from a template by FreeHTML5.co