A METHOD FOR THE ONLINE USE OF OFF-LINE DERIVED REMAPPINGS OF ITERATIVE AUTOMATIC TARGET RECOGNITION TASKS ONTO A PARTICULAR CLASS OF HETEROGENEOUS PARALLEL PLATFORMS
Jr. Budenske et al., A METHOD FOR THE ONLINE USE OF OFF-LINE DERIVED REMAPPINGS OF ITERATIVE AUTOMATIC TARGET RECOGNITION TASKS ONTO A PARTICULAR CLASS OF HETEROGENEOUS PARALLEL PLATFORMS, Journal of supercomputing, 12(4), 1998, pp. 387-406
Citations number
28
Categorie Soggetti
Computer Science Hardware & Architecture","Computer Science Theory & Methods","Engineering, Eletrical & Electronic","Computer Science Hardware & Architecture","Computer Science Theory & Methods
This study focuses on a particular application domain (iterative autom
atic target recognition tasks) and an associated specific class of ded
icated heterogeneous parallel hardware platforms. For the computationa
l environment considered, a methodology is presented for the on-line o
perating system to decide heuristically whether to perform a remapping
of the application onto the platform based on information generated f
rom input data by the application during execution. If the decision is
to remap, the operating system will be able to select a mapping, whic
h is appropriate for the given state of the application, from a stored
set of mappings that were previously derived with an off-line heurist
ic.