Total number of participants: 14
Number of different nationalities represented: 4
Total number of speakers: 1
Total number of talks: 1

Abstract:

High performance computers utilize lots of parallelism at different levels of vectorization, thread parallelism, message-passing between distributed memory architectures and even function off-loading by hardware accelerators. Large scale numerical simulations often offer lots of parallelism, which may be difficult to express in a high level programming language. A common abstract parallel programming style for numerical applications is proposed, which can be translated automatically into parallel code for one or a combination of common programming styles for these parallel architectures.

The slides of Prof. Zumbusch's guest lecture from today can be downloaded from here:

    http://heim.ifi.uio.no/~xingca/zumbusch08.ppt

The sourceforge address for Prof. Zumbusch's compiler, libraries and benchmark codes:

    http://parallel-for.sourceforge.net/

For more info about Prof. Zumbusch's publications, please visit
    http://cse.mathe.uni-jena.de/cgi-bin/bib/show.pl?author=Zumbusch