Computing Mode Selection using Software Defined Network


Computing Mode Selection in Cloud Computing introduces the latency in the network. The node at the lowest level of network has to wait for the process to finish which has been passed to the cloud for faster computation. Edge computing plays a vital role to avoid this delay in computation for every task as with fewer resources than cloud, fog computing can provide almost real-time computations to IoT node. We optimized this computing mode selection from local, fog and cloud computing.

This repository contains:

  • MATLAB code for the computation Model Selection using WOA in the Software Defined network
 Discuss Code



Computing Mode Selection using Software Defined Network started from the evaluation of cloud computing. The increase in IP traffic is due to the increase in the number of Internet-connected devices. Cloud computing and the Internet of Things (IoT) are two major technologies that have contributed a lot in this direction. With a reduction in the cost of sensing, computation and communication, more and more things are connecting to the Internet. These things are communicating with other connected things and humans.

Cloud computing concept started in 1950s with the evolution of mainframe computing. Wherein, multiple users were allowed to access a central computer through dumb terminals, whose sole function was to provide access to the mainframe.

The fog computing is the new invention which described as the intermediate layer among the cloud data and IoT devices or sensors. The concept of fog computing was initialized in the year 2012 by CISCO for the IIOT devices applications. Fog computing is reduced the several drawbacks of cloud computing and provided efficient latency in a stable manner. The fog computing system contains the traditional accessories like a router, base stations, servers, switches etc. and placed near about the IIOT devices. The fog computing improved the scalability, real-time interactions and mobility of the network IIOT system. Fog computing reduces computational cost, latency, power consumption, network traffic, etc. Yo

You are reading this at and can download the MATLAB code also.

Problem Statement

Traditional IoT architectures connect the sensing devices to the Internet and send the generated data to a cloud resource for processing. This methodology works well for the applications where strict delay is not a concern. Cloud is not an ideal resource for the applications which require real time responses. This is the reason, domains such as telecommunication, health care, real time control etc. process the data closer to the origin [1]. This computational paradigm is termed as fog computing.

Software Defined Network

Analogically, the evolution of software defined network (SDN) is the same as computer systems. Similar to this evolution, in SDN the network infrastructure has been divided into data, control and application plane. There are open standards for communication between these planes. Similar to the computer system, in SDN, applications at the application plane control the entire network infrastructure.

Software Defined Network

SDN Data Plane

The data plane contains the devices which control the data traffic flow. These devices communicate with control plane devices or SDN controllers using Open Flow (OF) protocol. In SDN paradigm, this interface is also called as South bound interface and devices as Open Flow switches. OF switches process incoming packets consulting the forwarding tables. These tables are configured by SDN controller using OF protocol. There are three kinds of these tables named as Flow table, Group table and Meter table.

SDN Control Plane

Similar to the functionality of the operating system in computers, SDN control plane can also be seen as Network Operating system (NOS) which maps the service requests made by application plane to OF complaint commands and provides information about data plane activity, topology and other statistics to application plane.

Control plane provides following functionality

  • Configure data plane devices as per application request.
  • Managing switch topology at data plane.
  • Collecting traffic statistics from OF switches.
  • Configure shortest path for a flow at data plane.
  • Manage notifications from data and application plane devices.
  • Provide security mechanisms.

SDN Application Plane

The application plane contains applications for various tasks. These applications may be used for managing wireless and mobile networks. Others may be used for monitoring and measurements of network traffic. Another application may be used for enforcing QoS rules on the data plane traffic. The network service abstraction layer exposes only necessary details to applications. This layer hides low level details of the data plane and provides high level APIs to interact with control plane

Proposed Work

This work is based on the work by Wang J. The proposed model contains three main layer cloud layer, terminal device layer and Fog computing layer. the cloud layer deal with the non real time applications, terminal layer uses for the industrial production and data transmission, and fog computing provided the real time based task computation by providing the SDN controller to optimize the CMS via WOA.

As we studied earlier in case IIOT environmental conditions the heterogeneous terminal devices and task are present. The computing model selection for the task process through the computing selectors was performed by terminal devices. The optimal computing mode selection is obtained by the terminal devices for the minimization of computation cost and time. We design the automatic computing mode selector by proposing the WOA optimization algorithm in the SDN controller. The optimal selection of computing mode is performed with Whale Searching food behavior as per WOA.

The local, cloud and fog are the three computing modes having different advantages and features. While implementing the WOA for the computing mode selection two categories of task unload and Offload task should be considered. The unloadable task is that which provided by the terminal devices present in the model like a calculator use in the smartphone. The task processed via edge server or cloud server is known as Offload task like server applications performed in the Smartphone.

In the first case, the terminal device processed a task so mode selection service request sends to the SDN controller. The response provided to the service request received all the information is transmitted to the SDN controller like task data, computation ability of the device, the content of data, observational amount and higher tolerance time. Similarly, the fog and cloud computing are sent their information like bandwidth range, transmission power and computation capacity to the SDN controller. The real time based tasks during cloud and fog computing achieved via WOA optimization algorithm processing within the SDN controller.

Among the three computing mode, only one is selected for the SDN controller. The computing mode takes fewer tasks execution time selected for the SDN controller processing.


The work implemented the computing mode selection for faster computation for several tasks. it optimized the distribution of tasks amongst cloud server, fog server and local server by WOA optimization. MATLAB code has been written for it.


There are no reviews yet.

Only logged in customers who have purchased this product may leave a review.