The simulator has been tested with Python-2.7. It may / may not work with Python-3.
The following packages are required to run the simulator:
- z3 v4.8.9 - 64 bit (Download from: https://github.com/Z3Prover/z3/releases)
- matplotlib (sudo apt install python-matplotlib)
- colorama (pip2 install colorama)
Colorama is used to stylize the output messages from the simulator. It is not required (raw messages will be printed) but is recommended.
Use the following command to see the options for configuring / running the simulator:
./simulator -h
usage: simulator.py [-h] [-p | -d] [-v] [-t {light,mixed,heavy}]
[-n {1,10,100,1000}] [-e {0,25,50,75}] [-r {0,r}]
[-m {4,8,16,32}] [-N {4,8,10,12}]
optional arguments:
-h, --help show this help message and exit
-p, --pristine collect pristine data
-d, --demo_mode illustrate the simulation steps with an example run
-v, --verbose specify output verbosity
-t {light,mixed,heavy}, --taskset_type {light,mixed,heavy}
type of taskset
-n {1,10,100,1000}, --num_of_tasksets {1,10,100,1000}
number of tasksets to generate for each util. point
-e {0,25,50,75}, --edge_probability {0,25,50,75}
edge probability in generated tasksets
-r {0,r}, --demand_type {0,r}
type of resource demand
-m {4,8,16,32}, --core_count {4,8,16,32}
number of cores 'm' for the simulation
-N {4,8,10,12}, --tasks_per_period {4,8,10,12}
max. number of tasks in any candidate-set
There is a demo mode of the simulator which illustrates the steps of the simulation using default parameters. To try that, run the simulator as follows:
./simulator -d