Scalability analysis asks how performance of a certain application
(or application class) behaves as the application problem size
increases and the parallel architecture executing it increases.
Scalability studies generally concern very specifically defined
applications, e.g. , FFT, global reductions, and many other precisely
defined operations [4]. Parallel and distributed discrete-event
simulation (PDES) is a critical technology for an important class
of very large complicated simulation models. However, with few
except,ions, the bulk of empirical work in PDES has been on relatively
small models. Furthermore, synchronization behavior is frequently
complicated, which makes it very difficult to analytically prove
anything about performance executing large models on large machines.
|