Cluster Sweep
Use the Cluster Sweep study step when you want to study to solve several models in parallel where each model has a different set of parameters.
Syntax
model.study(stdname).create(fname, "ClusterSweep");
model.study(stdname).feature(fname).set(pname,value);
Description
Study step.
The following properties are available.
filled | sparse
on | off
Probes to use when probesel=manual.
all | none | manual
on | off
filesystem | database | sameascurrentmodel
filesystem will store the data on the file system in the configured batch directories.
database will store the data in the same database and branch as where the original model is stored.
sameascurrentmodel will store the data where the original model is stored, which can be either the file system or a database.
filesystem | database | sameascurrentmodel
filesystem will store the models on the file system in the configured batch directories.
database will store the models in the same database and branch as where the original model is stored.
sameascurrentmodel will store the models where the original model is stored, which can be either the file system or a database.
on | off
on | off
on | off
general | whpc2008 | sge | slurm | pbs | lsf | none
on | off
on | index
on | off
on | off
on | off
on | off
on | off
none | scp | user
on | off
none | ssh | user
native | windows | linux
ssh | putty | user
on | off