Members
Overall Objectives
Research Program
Application Domains
Highlights of the Year
New Software and Platforms
New Results
Bilateral Contracts and Grants with Industry
Partnerships and Cooperations
Dissemination
Bibliography
XML PDF e-pub
PDF e-Pub


Section: Partnerships and Cooperations

International Initiatives

Collaboration with Colorado State University

Compsys had always kept strong connections with Colorado State University (CSU):

Polyhedral Community

In 2011, as part of the organization of the workshops at CGO’11, Christophe Alias (with Cédric Bastoul) organized IMPACT’11 (international workshop on polyhedral compilation techniques, http://impact2011.inrialpes.fr/). This workshop in Chamonix was the very first international event on this topic, although it was introduced by Paul Feautrier in the late 80s. Alain Darte gave the introductory keynote talk. After this successful edition (more than 60 people), IMPACT continued as a satellite workshop of the HIPEAC conference, in Paris (2012), Berlin (2013), Vienna (2014). Alain Darte was program co-chair and co-organizer of the 2015 edition in Amsterdam, and Tomofumi Yuki of the 2016 edition in Prague.

The creation of IMPACT, now the annual event of the polyhedral community, helped to identify this community and to make it more visible. This effort was complemented by the organization by Alain Darte of the first school on polyhedral code analysis and optimizations (http://labexcompilation.ens-lyon.fr/polyhedral-school/). A second polyhedral school (https://mathsinfohpc.sciencesconf.org), more open, because involving themes and researchers from numerical analysis (users of HPC), was organized in 2016 by Alain Darte (for the compiler side) and Violaine Louvet (for the HPC side). See details in Section 10.1.

Alain Darte also manages two new mailing lists for news (polyhedral-news@listes.ens-lyon.fr) and discussions (polyhedral-discuss@listes.ens-lyon.fr) on polyhedral code analysis and optimizations. Tomofumi Yuki is involved in the development of PolyBench (http://sourceforge.net/projects/polybench), a suite of kernels used for illustrating polyhedral optimizations. He is also developing PolyApps, a set of larger applications to evaluate the gap between kernels and “real” applications, see more details in Section 7.7.