Select Profiles (under Darwin) in the left list.
Click 
.
In the Profile name field, enter a name for the profile.
From the Run searches pulldown, select how searches are to be run when this profile is used:
– locally
– on remote host as foreground process
– on remote host as grid job
Set the maximum Number of models to allow to be executed in parallel.
In the Path to nmfe script field, type the full path to the script or click 
to browse to the location.
In the Path to Rscript field, type the full path to the Rscript executable or click to browse to the location.
In the SSH login command field, type the command for connecting to the host.
In the Remote mount location field, enter/edit the name a folder on the remote host that you have mounted as a local drive, typically /home/username (e.g., /home/smith).
In the Local mount location field, enter/edit the drive-letter on the local system that corresponds with the remote host path defined in the previous field, such as X: or V:.
In the Path to Python interpreter field, type the path to the Python interpreter.
In the Command to enqueue a search job field, type the command to use when submitting an entire search job to the grid queue.
In the Command to enqueue an individual run field, type the command to use when submitting individual runs to the grid queue.
In the Submitted job id regular expression field, type the information to display when a job is submitted.
In the Command to poll finished runs field, type the command to retrieve finished jobs from the grid controller.
In the Polled job id regular expression field, type the information to display when a job is retrieved.
In the Poll interval field, set the number of seconds between checks for finished jobs.
In the Command to delete search related jobs field, type the command to use to delete all unfinished jobs related to a search that is stopped.
Clicking “?” opens the pyDarwin documentation page with more details about that option in a web browser.