 |
 |
 |
 |
 |
 |
I am Research Scientist (CR1) at INRIA, in the Moais team, Grenoble.
My current research activities focus on High Performance Interactive Computing (HPiC). Applications are related to scientific visualization, virtual reality and telepresence. The goal is to design and develop algorithms and software layers to ease the development of large interactive applications taking advantage of the resources offered by machines involving complex input devices (cameras), numerous computing resources (mixing CPUs and GPUs), and multiple outputs (projectors).
Multi-core architectures (CPUs and GPUs) offer an unique opportunity for HPiC. I recently focused on algorithms to take advantage of these architectures for scientific visualization (cache oblivious mesh layouts and layout consistent parallel algorithms) with Marc Tchiboukdjian, but also on supporting multi-CPUs and multi-GPUs execution of the SOFA physics engine with Everton Hermann.
I am co-steering the Grimage platform with Edmond Boyer.
It is an innovative platform gathering a PC cluster, several digital cameras and displays.
Grimage is used to experiment novel interactive applications, like telepresence or interactive scientific applications,
requiring the power of tens of processors provided by the local cluster or through the grid. See the two succesfull demos
at Siggraph 2007 and Siggraph 2009.
Grimage also led to the creation of the succesful 4DViews Company.
FlowVR is the
middleware we develop to ease development and execution of iteractive
applications distributed on clusters and grids. FlowVR relies on a hierarchical component model and a data-flow paradigm.
FlowVR enables parallel code coupling and advanced coherency
control between data flows. Amongst others, Grimage
applications are developped with FlowVR.
Ph.D. students I really enjoy(ed) to work with:
Initial works in virtual reality led to the development of the Net Juggler and SoftGenLock
Libraries. Net Juggler distributes graphics rendering on a PC
cluster. SoftGenLock enables active stereo on a Linux PC cluster with
commodity graphics cards.
From 1999 to 2001 I was Assitant Professor at LIFO, Université
d'Orléans. I taught cryptography and network security, computer
architecture, parallel programming, networking, object programming and
operating systems classes.
I worked almost two years (98-99) at Iowa State University
on parallel computer performance evaluation and taught
few calculus classes. Research work was done with
Pr. Glenn R. Luecke in close collaboration with Cray and
SGI. I had the unique opportunity to run codes with very
large numbers of processors on machines like the Blue
Mountain cluster of Origin 2000 located at Los Alamos,
or a 1500 processors T3E-1200 located at Eagan,
Minnesota (at the time 1500 processors was really a lot).
I obtained a Computer Science Ph.D. from Université d'Orléans in
1997. Research work aimed at developing structured
synchronizations and communications models for message
passing parallel programming. With Emmanuel
Melin, Xavier Rebeuf and Bernard Virot, we explored many aspects of this topic, going
up to automatic parallelization and cost models.
I also did some research about learning rules for artificial neural networks
with Mirta B. Gordon and Bernard Virot.