Global Grid
Forum - Amsterdam, March 2001
Targets for HPC in next
decade
lCapture more physics in the simulation of
complex systems
lComplex systems are characterized by multiple
time and/or length scales
lNot easy to capture multiple scales in one
code
lCode coupling for multi-physics applications is viable alternative in some conditions
lThis leads naturally to computational grids
lHeterogeneous algorithms map naturally to heterogeneous grids.