I have some questions similar to Managing multiple anylogic simulations within an experiment, however in more detail:
We'd like to use AnyLogic simulation (especially Discrete Event Simulation) with a multi-node MonteCarlo approach, using variable parameter initializations. Each single simulation run should trigger a run of the jar file exported from our AnyLogic project. The overall experiment should be started by command line on Ubuntu Linux.
For this we need some non-interactive mode of the simulation in order to boost the performance as follows:
How can we achive this with AnyLogic?
Additionally, we also like to show the animations for a single simulation run, when a user provides concrete parameter values. For this, we could e.g. install AnyLogic on the Linux server or on a virtual machine running on top.
Thanks in advance
Luma
You can use the command getEngine().setRealTimeMode(false)
to run a simulation exp as fast as possible.
If you use a boolean parameter on main "timeMode", you can govern it from the command line input to decide if it should run as fast as possible or not.
Else, to turn off animation completely, the easiest way is to just use a ParameterVariation experiment, set it to run in "Freeform" mode with 1 iteration. This will boost your performance a lot.