Progress...

Yesterday I added some minimal timing code to my experiment to get (theoretical) nanosecond timing. I used to rely on the printed output of mvn, but I don't like that this included the forking of the java executable.

I also changed the simulation script that tries out different queueing strategies by passing different XML files to the simulator. It currently outputs simple CSV data, containing the nanosecond timestamps gathered from standard output of the simulator.

I also implemented a CalendarQueue (from here on called CQ), based on the one used by Ptolemy II, in the simulator. Porting it over was a bit harder than expected, since the Ptolemy II CQ wasn't generic ("templated" on the event), used all sorts of internal Ptolemy II classes like Time and most of all, it uses a "smart" way of structuring the CQ. This structuring takes into account when the contained events fire to structure the CQ.