Michel Schanen

Research

My focus is on the efficient computation of adjoints in a large computer cluster setting. The de facto distributed parallelization standard is the message passing interface (MPI). The adjoint algorithmic differentiation model is directly linked to a data flow reversal problem, thus implying a reversal of all message passing communications.

The first goal has been the development of an adjoint MPI library, providing a generic reversal mechanism for any MPI enabled program in combination with an arbitrary algorithmic differentiation tool.

Short CV

cv.pdf

Funding

This research is supported by the Fond National de la Recherche of Luxembourg under grant PHD-09-145.

Publications

Michel Schanen, Michael Förster, Boris Gendler, and Uwe Naumann. Compiler- based Differentiation of Numerical Simulation Codes. In ICCGI 2011, The Sixth International Multi-Conference on Computing in the Global Information Technology , pages 105-110. IARIA, 2011.

M. Schanen, U. Naumann, L. Hascoët, and J. Utke. Interpretative Adjoints for Numerical Simulation Codes using MPI. In Procedia Computer Science, pages 1819–1827. Elsevier, 2010.

Michel Schanen, Michael Förster, and Uwe Naumann. Second-Order Algorithmic Differentiation by Source Transformation of MPI Code. In Rainer Keller, Edgar Gabriel, Michael Resch, and Jack Dongarra, editors, Recent Advances in the Message Passing Interface, volume 6305 of Lecture Notes in Computer Science, pages 257–264. Springer Berlin / Heidelberg, 2010.

Contact Details

Michel Schanen  
Software & Tools for Computational Engineering 
RWTH Aachen University  
Office Phone: ++49-(0)241-80-20129  
Email: schanen at really no spam please stce dot rwth minus aachen dot de  
Jabber: Michel.Schanen at jabber.rwth-aachen.de