Mininet Network Simulator Performance Analysis

For the simulation of SDN networks, Mininet offers a robust platform. Assessing major network metrics such as packet loss, throughput, and latency is most significant while carrying out a performance analysis process. To configure and examine the performance of the network in Mininet, we suggest an important instruction:

Procedures for Performance Analysis

1. Install Mininet

In the beginning, you need to make sure that the Mininet is appropriately installed on your system.

sudo apt-get update

sudo apt-get install mininet –y

2. Develop a Network Topology

On the basis of your Mininet network topology, a Python script has to be developed. As an instance, a script for basic linear topology includes:

# linear_topology.py

from mininet.net import Mininet

from mininet.node import OVSController

from mininet.topo import LinearTopo

from mininet.cli import CLI

from mininet.log import setLogLevel

def linear_topology():

    # Set up a linear topology with 3 switches

    topo = LinearTopo(k=3)

    net = Mininet(topo=topo, controller=OVSController)

    net.start()

    CLI(net)

    net.stop()

if __name__ == ‘__main__’:

    setLogLevel(‘info’)

    linear_topology()

3. Installation of Network Traffic Tools

For latency assessment, install ping, and iperf for throughput assessment.

# Install iperf

sudo apt-get install iperf –y

4. Execute the Topology

Execute the script that you have developed for your Mininet topology.

sudo python3 linear_topology.py  

5. Throughput Assessment with iperf

• Start iperf Server: Initiate the iperf server on host h1.

mininet> h1 iperf –s

• Start iperf Client: To assess throughput, initiate the iperf client on host h2.

mininet> h2 iperf -c h1

• Output Instance:

Client connecting to h1, TCP port 5001

TCP window size: 85.3 KByte (default)

 [  3] local 10.0.0.2 port 54052 connected with 10.0.0.1 port 5001

[ ID] Interval       Transfer     Bandwidth

[  3]  0.0-10.0 sec  11.2 GBytes  9.63 Gbits/sec

6. Latency Assessment with ping

Among hosts, assess the network latency with the aid of ping.

• Ping Host: conversely, ping host h1 from host h2.

mininet> h2 ping -c 10 h1 

• Output Instance:

PING 10.0.0.1 (10.0.0.1) 56(84) bytes of data.

64 bytes from 10.0.0.1: icmp_seq=1 ttl=64 time=1.07 ms

64 bytes from 10.0.0.1: icmp_seq=2 ttl=64 time=1.02 ms

64 bytes from 10.0.0.1: icmp_seq=3 ttl=64 time=1.11 ms

64 bytes from 10.0.0.1: icmp_seq=4 ttl=64 time=1.04 ms

64 bytes from 10.0.0.1: icmp_seq=5 ttl=64 time=1.08 ms

— 10.0.0.1 ping statistics —

10 packets transmitted, 10 received, 0% packet loss, time 9001ms

rtt min/avg/max/mdev = 1.02/1.10/1.19/0.07 ms

7. Packet Loss Analysis

• Create Traffic: Create traffic with the help of iperf for simulating packet loss.

mininet> h1 iperf -s

mininet> h2 iperf -c h1 -u -b 10M

• Evaluate Packet Loss:

mininet> h2 ping -c 10 h1  

In the ping result, examine the packet loss percentage.

8. Flow Analysis and Visualization

To examine the flow of the network, employ inherent tools of Mininet or Wireshark.  

Capture Traffic with tcpdump:

mininet> h1 tcpdump -i h1-eth0 -w h1_traffic.pcap  

Visualize in Wireshark:

• In Wireshark, open h1_traffic.pcap.
• Packet information has to be studied. Then, visualize flow details.

Outline of Performance Metrics

• Throughput: Use iperf to assess throughput.
• Latency: Evaluate latency with the support of ping.
• Packet Loss: From ping outcomes, assess the packet loss.
• Flow Analysis: Employ Wireshark or tcpdump to examine flow.

Documenting Performance Analysis

Create an explicit document by encompassing the following aspects:

• Network Topology: Include explanation or design of your Mininet topology.
• Tools Employed: The tools that you have employed for the assessment process must be defined.
• Performance Metrics:
  • Consider Flow Analysis, Packet loss, Latency, and Throughput.
• Discoveries and Perceptions:
  • In terms of various contexts, compare the metrics.
  • It is approachable to detect any abnormalities or problems.

How can I run multiple SDN controllers Floodlight Ryu ODL etc for a topology in Mininet?

Appropriate arrangements and settings are needed to execute several SDN controllers like OpenDaylight (ODL), Ryu, and Floodlight, especially for a one Mininet topology. Based on this execution, we provide a procedural guideline in an explicit manner:

Outline

1. Install Controllers: It is important to make sure that the required controllers such as OpenDaylight, Ryu, and Floodlight are properly installed on your system.
2. Set up Controllers: For every controller, arrange various protocols and ports.
3. Develop a Mininet Topology: Construct a Mininet topology. Then, controllers have to be allocated to particular switches appropriately.

Procedural Instruction:

Step 1: Install SDN Controllers

Initially, you should ensure whether the controllers like OpenDaylight, Ryu, and Floodlight are installed in an appropriate manner.

Floodlight

# Install Java

sudo apt install openjdk-11-jdk

# Clone and build Floodlight

git clone https://github.com/floodlight/floodlight.git

cd floodlight

./gradlew shadowJar

# Start Floodlight (default port: 6653)

java -jar target/floodlight.jar

Ryu

# Install Ryu

pip install ryu

# Start Ryu controller (default port: 6653)

ryu-manager –ofp-tcp-listen-port 6654 ryu.app.simple_switch_13

OpenDaylight

# Download and extract OpenDaylight

wget https://nexus.opendaylight.org/content/repositories/public/org/opendaylight/integration/distribution-karaf/0.10.2-Helium/distribution-karaf-0.10.2-Helium.tar.gz

tar -zxvf distribution-karaf-0.10.2-Helium.tar.gz

cd distribution-karaf-0.10.2-Helium

# Start OpenDaylight

./bin/karaf

Step 2: Develop a Mininet Topology

After that, a Mininet topology script (multi_controller-topology.py) has to be developed, which specifies the network in a clear way:

# multi_controller_topology.py

from mininet.net import Mininet

from mininet.topo import Topo

from mininet.node import RemoteController, OVSSwitch

from mininet.cli import CLI

from mininet.log import setLogLevel

class MultiControllerTopo(Topo):

    def build(self):

        # Create switches

        s1 = self.addSwitch(‘s1′, protocols=’OpenFlow13’)

        s2 = self.addSwitch(‘s2′, protocols=’OpenFlow13’)

        s3 = self.addSwitch(‘s3′, protocols=’OpenFlow13’)

        # Create hosts

        h1 = self.addHost(‘h1’)

        h2 = self.addHost(‘h2’)

        h3 = self.addHost(‘h3’)

        h4 = self.addHost(‘h4’)

        # Add links between switches and hosts

        self.addLink(h1, s1)

        self.addLink(h2, s1)

        self.addLink(h3, s2)

        self.addLink(h4, s3)

        # Add links between switches

        self.addLink(s1, s2)

        self.addLink(s2, s3)

        self.addLink(s3, s1)