Deployments, Tests, and Datasets of Free5GC and Open5GS

1. Using the Datasets from Our Experiments:

1. Choose the core network you would like to use the datasets with:

• For Free5GC:

cd Free5GC/Dataset

• For Open5GS:

cd Open5GS/Dataset

2. Select the experiment you would like to explore:

• For Decrement_Test:
• 10 rounds of a test with an initial time interval of 4600ms, where the time is decremented by 500ms between 10 connection tests for 100 UEs:

cd Decrement_Test

• For Division_Test:
• 10 rounds of a test with an initial time interval of 51200ms, where the time is divided by a factor of 2 between 20 connection tests for 100 UEs:

cd Division_Test

• For Parallel_Test_100:
• 10 rounds with a constant time interval of 100ms between connection tests for 100 UEs:

cd Parallel_Test_100

• For Parallel_Test_10000:
• 10 rounds with a constant time interval of 10000ms between connection tests for 100 UEs:

cd Parallel_Test_10000

3. Select the type of data for analysis:

• For Tester:

  • 10 CSV files containing tester logs from our experiments.

• For timestamps.txt:

  • 10 timestamps collected from our experiments to be used for metric collection.

2. Running the scripts for new tests

1. Choose the core network you would like to test:

• For Free5GC:

cd Free5GC

• For Open5GS

cd Open5GS

2. Update the namespace in the YAML deployment files to match your configuration:

1. Go to Deployment directory

cd Deployment

2. Update this field in all YAML files:

namespace: your_namespace

3. Make the scripts executable using the chmod command:

cd ../Data/
chmod +x connection_test.sh
chmod +x capture_and_parse_logs.sh

4. Ensure that you have kubectl installed and are running in root mode:

sudo su
kubectl --version

5. Modify the initial variables in connection_test.sh to suit your requirements:

• replicas: Specifies the number of testers.
• namespace: Indicates where kubectl commands will be executed.
• sleep_time: Defines the duration required to execute one round of the experiment.

6. Choose the workload test you would like to perform:

1. For constant time interval between connections:

./connection_test.sh parallel [number_of_ues_per_tester] [delay] [rounds]

• For example: 10 rounds with a constant time interval of 1000ms between connection tests for 100 UEs:

./connection_test.sh parallel 100 1000 10

2. For exponential decreasing time between connections:

./connection_test.sh division [number_of_ues_per_tester] [start_delay] [interval] [factor] [rounds]

• For 10 rounds of a test with an initial time interval of 51200ms, where the time is divided by a factor of 2 between 20 connection tests for 100 UEs:

./connection_test.sh division 100 51200 20 2 10

3. For linear decreasing time between connections:

./connection_test.sh decrement [number_of_ues_per_tester] [start_delay] [interval] [factor] [rounds]

• For 10 rounds of a test with an initial time interval of 4600ms, where the time is decremented by 500ms between 10 connection tests for 100 UEs:

./connection_test.sh decrement 100 4600 10 500 10