spider-pcn / spider_omnet Goto Github PK

This is a simulator that models the flow of transactions through a payment channel network.

To setup, first follow the omnet installation instructions from Omnet Installation. Based on your version of omnet, add this line to your default shell rc file.

export OMNET="omnetpp-5.6"

Next, clone this repo in the samples directory of the omnet installation and then run the following:

cd samples/spider_omnet/
make

To use the Omnet IDE, run omnetpp which opens the IDE allowing you to click on the simpleNet.ned file and run the OmnetSimulation from there. Once you have completed installation, navigate to the folder with setenv (should be first level inside the omnet simulation folder) and run the following to save rebuilding time.

. setenv
omnetpp

Open the simpleNet.ned file and click the green run button to run the simulation.

This software needs Python3 and the following python modules can be installed via pip: networkx, matplotlib

It also requires Gurobi. Once installed, navigate to the Gurobi installation folder and run python setup.py install to get the associated gurobipy module.

To run from the command line, either use the run_exp.sh script or by running simpleNet using the command

./spiderNet -u Cmdenv -f omnetpp.ini -c simpleNet -n .

If you use the run_exp.sh script, the plots can be found in the scripts/figures directory.

Click the button to run the simulation. For just the purpose of collection statistics, you can use the EXPRESS setting (triple arrow) to run through the simulation. Once the simulation has terminated, close the simulation window.

Inside the omnet IDE, open the results folder. The statistics of the most recent run will be saved in the "simpleNet-#0.vec" file. To view the file, rename the .vec file to something unique, for example "figure4b-time10.vec". Double click the renamed file, which will prompt you to generate a .anf file. Click Finish.

In the .anf file, click Browse Data on the bottom toolbar. The individual nodes are in the expanded view under the folder "simpleNet: #0". Expand a node to see the statistics collected for that node. Right click on a node, folder, or statistic to plot it.

• The third white drop down menu next to the blue arrow can filter by statistic.
• If you want to filter by more than one statistic, use the funnel button next to the blue arrow, and query for example:

name("numAttemptedPerDest_Total(dest *") name("numCompletedPerDest_Total(dest *")

• For aesthetic graphs, you need to right click -> properties, add titles and axis under "Titles", check display legend under "Legend"