Skip to content

AdvancedNetworkingSystems/wcn_emulator

master
Switch branches/tags
Code

Latest commit

 

Git stats

Files

Permalink
Failed to load latest commit information.
Type
Name
Latest commit message
Commit time
 
 
community_networks_analysis @ ec05b0d
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

NePA TesT is a front-end for mininet which setups experiment on a given (networkx) network

Setup

NePA TesT requires a patched version of Mininet. Please refer to: NePA wiki for the detailed setup.

Usage

To run a test you need a configuration file (to be placed in conf/) and the test class code (to be placed in test_code/). Then run $>python wcn_emulator -f <conf_file> -t where identifies a test case in <conf_file>. Each test case must specify an implementation of MininetTest through the argoment "testClass"/ Each MininetTest class has the runTest method which starts what the test needs and then waits for completion. This waiting can be implemented with self.wait() function which sleeps for the desired amount of time and optionally logs resource usage.

For example:

  • self.wait(self.duration) # will wait till the end
  • self.wait(self.duration, log_resources={'cpu': 'cpu_usage'}, 2) # will wait till the end of the test while sampling every 2 seconds the cpu load and saving the results in a file called 'cpu_usage'.

Other than 'cpu', also 'net' and 'mem' can be logged; 'mem' makes NePA TesT logging the amount of available RAM (free or released) and the amount of swap memory used; 'net' makes instead logging both the fraction of received packets over sent packets and the fraction of received bytes over sent bytes since the beginning of the test. It is important to always specify the destination file of the logs (even the same for all the resources).

Test

NePA TesT stores test files in the Test folder. Tests follows the pytest approach, to run them simply install the pytest module and run $>python -m pytest in the root folder

Troubleshooting

If something fails system may remain unstable. In this case try typing: $> sudo mn -c

How to

Use a custom delay distribution for the links

Through network_builder.py is possible to set some parameters of the links, e.g., loss, bandwidth, mean delay, delay jitter and delay distribution.

Available delay distributions are placed in /usr/lib/tc/ and the defaults one are normal, pareto, paretonormal and experimental.

Given sample delay values is possible to build a custom distribution and use it in mininet. The following is an extraction from this netem article.

  1. git clone git://git.kernel.org/pub/scm/linux/kernel/git/shemminger/iproute2.git
  2. compile these netem utilities: cd iproute2/netem gcc -o maketable maketable.c -lm gcc -o stats stats.c -lm
  3. fill data.txt with samples from the distribution (in ms), e.g., ping -c10 google.com | grep -Eo 'time=[0-9.]+'| grep -Eo '[0-9.]+' > data.txt
  4. build a netem distribution ./maketable data.txt > mydist.dist
  5. put mydist.dist in /usr/lib/tc/
  6. compute mean and standard deviation (sigma) ./stats data.txt
  7. use the following configuration parameters: link_mean_delay = ms link_delay_sd = ms link_delay_distribution = mydist

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages