On Wed, 24 Dec 2008, Kyle Spaans wrote:
Take that as you will, but for me it only means that Prof. Dongarra is only tengentially related to beowulf through NETLIB FORTRAN code. And thusly, probably is not a ``mad scientist'' of beowful fame. ;-)
Dongarra was one of the primary people involved in the original development of PVM, which was the original cross-platform parallel/network programming support package. MPI had very different roots and was not originally designed to support network-based parallel computing. So in a sense, Dongarra was one of the real inventors of the commodity cluster. Post PVM, it was EASY to take existing workstations on a TCP/IP network and write parallel code, and the beowulf design was nearly inevitable as soon as Linux matured to where it could support it. Dongarra is also one of the people who intiated the ATLAS project, a linear algebra package that can yield as much as a factor of 2-3 performance edge over non-tuned linear algebra packages. For people with LA-intensive code, that's like doubling or trippling the size of their clusters. A bit of a mad scientist, sure. Or at least, a Very Smart Guy. rgb
_______________________________________________ Beowulf mailing list, Beowulf@beowulf.org To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf
Robert G. Brown http://www.phy.duke.edu/~rgb/ Duke University Dept. of Physics, Box 90305 Durham, N.C. 27708-0305 Phone: 1-919-660-2567 Fax: 919-660-2525 email:r...@phy.duke.edu _______________________________________________ Beowulf mailing list, Beowulf@beowulf.org To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf
