HR: 1340h
AN: ED13C-1155    [Abstracts]
TI: Parallelizing a peanut butter sandwich
AU: * Quenette, S M
EM: steve@vpac.org
AF: Victorian Partnership for Advanced Computing (VPAC), 110 Victoria Street, PO Box 210, Carlton South, Melbourne, VIC 3053 Australia
AB: This poster aims to demonstrate, in a novel way, why contemporary computational code development is seemingly hard to a geodynamics modeler (i.e. a non-computer-scientist). For example, to utilise comtemporary computer hardware, parallelisation is required. But why do we chose the explicit approach (MPI) over an implicit (OpenMP) one? How does this relate to the typical geodynamics codes. And do we face this same style of problems in every day life? We aim to demonstrate that the little bit of complexity, fore-thought and effort is worth its while.
DE: 0500 COMPUTATIONAL GEOPHYSICS (3200, 3252, 7833)
DE: 0845 Instructional tools
DE: 3200 MATHEMATICAL GEOPHYSICS (0500, 4400, 7833)
DE: 8100 TECTONOPHYSICS
SC: Education and Human Resources [ED]
MN: Fall Meeting 2005