Home → Documentation Home → Running a Simulation → Runtime GUIs → Monte Monitor |
---|
Monte Monitor (hereafter referred to as MM) is a graphical user interface that allows users to view and modify the states of slaves in a Monte Carlo simulation.
MM can be launched from the command line via:
''' ${TRICK_HOME}/bin/mm [options] '''
Pass --help
for a description of available options. For additional launching options, see
"Automatically Launching Applications".
The GUI pictured below may have a different look and feel based on the architecture of the machine on which it is running, but the functionality will remain the same.
The progress bar displays a visual and textual representation of how far the Monte Carlo has progressed. The format is
<resolved runs> / <total runs> (<percent resolved>)
.
The slave table displays information on each of the slaves. From left to right, this information is:
Ready
state. Since Monte Carlo cannot determine if slaves spawn correctly, slaves
that fail to spawn will remain in this state for the duration of the Monte Carlo.
Ready
state.
Stopped
state.
Running
state.
Stopping
state.
The new slave field allows the user to add a new slave by machine name while the Monte Carlo is running. Note that slaves
that fail to spawn for any reason (a bad machine name, for instance) will nevertheless appear in the slave table with a
status of Initializing
.
The connection panel displays host and port information when MM is connected to a simulation, and allows the user to specify these parameters when disconnected.
Slaves can be started and stopped while the Monte Carlo is running via the Monte Carlo
and
Slaves
menus. Selecting a command from the Monte Carlo
menu is equivalent to applying the same
command from the Slaves
menu to all slaves in the slave table. Stopping a slave will prevent any further
runs from being dispatched to it after its current dispatch completes. Starting a slave will resume dispatches to it.