1 of 11

ITEC626: INFORMATION SYSTEMS INFRASTRUCTURE��

Software Defined Networking SDN

2 of 11

Outline

Definition of SDN 5. How SDN Works
Evolution of SDN 6. Importance of SDN
SDN Layers Diagram 7. Challenges of SDN
SDN Layers Discussion 8. Applications of SDN

3 of 11

Definition of SDN

SDN is a technology that is intended to make networks more adaptable and require less maintenance.
SDN streamlines governance by separating the control plane from the data conveyance function in independent networking devices, allowing it to be managed from a central location.

(Xie et al., 2018)

4 of 11

Evolution of SDN

The beginnings of Software Defined Networking (SDN) can be traced back to 1995, when Sun Microsystems published the Java programming language.
AT&T's Geoplex was one of the first and most notable SDN initiatives.
In 2008 SDN technology was advanced to yield OpenFlow protocol
The OpenFlow protocol serves as a fundamental building block for the development of SDN solutions.

SDN has been in operation for more than a decade. In the early days of software-defined networking, many people believed that gleaming software-defined networks would eventually replace tightly tied, vertically integrated networking equipment (Valdivieso Caraguay et al., 2014). SDN principles were implemented to ease deployment and management of public switched telephone networks. It was necessary to separate the control, and data planes, first implemented in public switched telephone networks long before similar architecture was implemented in data networks (Valdivieso Caraguay et al., 2014). OpenFlow is only one of the original SDN canons, but it is a critical component because it was the catalyst for the change in networking software. It defined a programmed network protocol that might assist in managing and directing traffic among networking equipment, regardless of which vendor manufactured the underlying wired network at the time.

5 of 11

SDN Layers�

The application, control, and infrastructure layers are the three layers that make up a standard model of software-defined network architecture. As may be expected, the SDN application layer incorporates the conventional network applications and functionalities such as detection techniques, load balancing, and firewalls (Haleplidis et al 2015). The behavior of the data plane is managed by a controller in a network structure, whereas an application in a software-defined network manages the behavior of the data plane. Networks are segmented into three distinct levels under the control of the SDN architecture; for example, applications engage with the control layer through northbound APIs, and the control layer connects with the data plane through southbound APIs (Haleplidis et al 2015). . The control layer is referred to as the brain of the SDN architecture. The SDN controller software, which is centralized, provides the intelligence for this layer. This controller, which is housed on a server, is in charge of managing policies and the flow of traffic throughout the network. The infrastructure layer of a network is made up of the physical switches that connect the nodes.

6 of 11

How SDN Works

The SDN initiates separation of functions value by reducing the number of interdependencies.
The centralized controller sends the interdependencies as bandwidth rules to the switch.
The switch gives the controller details on the network it manages, as it requests the controller for instructions as appropriate.
To ensure that all packets intended for the same server are treated uniformly, the switch routes them along the same route. In order to implement the virtualization part of SDN

SDN is a collection of technologies that are coordinated together. Among the essential components are network virtualization and automation through well-defined APIs. The separation of functions adds value by reducing the number of interdependencies. Assume you have a packet that arrives at a network switch and that the rules incorporated into the switch's custom firmware instruct the switch where to route the packet (Xie et al., 2018). It is the centralized controller that sends these bandwidth rules to the switch. It also gives the controller details on the network it manages, as the switch, also known as a data plane device, requests the controller for instructions as appropriate. To ensure that all packets intended for the same host are treated uniformly, the switch routes them along the same route. In order to implement the virtualization part of SDN, a virtual overlay must be created on top of the physical network, which is a logically independent network.

An end-to-end overlay can be used to segment network traffic and improve performance (Valdivieso Caraguay et al., 2014). Consequently, users can abstract away from the underlying network and the network itself. It is particularly beneficial for service providers and operators with multi-tenant cloud resources and cloud solutions. They may set up a separate virtual network with particular policies for each client, reducing costs and increasing efficiency. The technologies of Network Function Virtualization (NFV) and Software-Defined Networking (SDN) are quite complementary. Network functions virtualization (NFV) abstracts network services from their underlying hardware. The availability of a large number of physical devices that perform specific functions, such as load balancers, routing, and switching devices, WAN accelerators, or content filters, is becoming more common these days. When it comes to delivering new network services, service providers look to network virtualization (NFV) as a solution

7 of 11

Importance of SDN

SDN enhances network consumer support connectivity, communication and safety of files sharing
It enables firms to use software and hardware from various suppliers by having tailored infrastructure
SDN accelerates integration of new applications and business models

Because of the abstraction from hardware and all of the limits that a hardware-bound network historically had, SDN technologies may construct networks that can promote innovation, provide new services, lower Capex and Opex, and provide fluidity and flexibility, among other benefits. Through SDN, organizations can facilitate innovation by developing new types of apps, services, and business models in a fraction of the time and with less fuss than they could previously. Organizations can create new services and revenue streams through dynamic applications and a more streamlined workflow process. Network functions can be performed on less expensive off-the-shelf hardware, resulting in a reduction in capital expenditure. Enterprises can lower their out-of-pocket expenses on information technology services by enabling automation and algorithm control through software-defined networking (SDN). These approaches and technologies allow enterprises to rapidly deploy new applications, services, and fluid and flexible infrastructure to meet their changing requirements.

8 of 11

Challenges of SDN

Meeting the demands of a rapidly growing market
Adequately dealing with variable, real-time change
Combining the service context in the integration process.

Rapid on-demand expansion creates a threat to monitoring systems. The inevitable increase in demand for new computing, network, and storage resources in software-defined infrastructure poses a threat to monitoring platforms. These systems must scale up in terms of monitoring capabilities to keep up with the continuous expansion of infrastructure. They may become oversubscribed very rapidly if they cannot add more capacity on demand (Xie et al., 2018). This will result in performance visibility gaps. In contrast to legacy infrastructure, we can have numerous overlay topologies running on top of the virtual network in the SDN world, which is not possible with legacy architecture. The ability to automate the provisioning of new converged infrastructures in minutes and influence multiple devices simultaneously is a game-changer, especially when compared to today's relatively static systems, which rely on human settings (Valdivieso Caraguay et al, 2014). Addressing dynamic real-time change Due to SDN, new computing, network, and storage devices and features are made accessible for usage almost instantaneously. When checking to see what's new, these dynamic, real-time changes can result in substantial gaps in invisibility in the future. Having service context is now considered standard practice. A consequence of this is that network performance monitoring must be able to listen to specific customers or tenants to be effective in its monitoring.

9 of 11

Applications of SDN

Network Intelligence and Monitoring
Applications on regulatory and compliance-related regulations
Can be used in high performance application such as CAD
Security Services

Network Intelligence and Monitoring are two terms that are used interchangeably. Modern software-defined networking (SDN) technologies assist in the abstraction of one of the most crucial layers within the data center: the network. Network infrastructures have become much more complicated, and they must manage far more data than they did in the past. This means that being aware of the flow of information through your environment is crucial.

Applications with a focus on compliance and regulation. Major cloud companies are now able to store and deal with workloads that must adhere to regulatory requirements. Organizations now can expand designs that were previously restricted by rules into distributed settings and the cloud, allowing them to innovate more quickly.

Applications with a high level of performance. We are witnessing an explosion in the number of different sorts of application technologies. Virtualization has enabled the delivery of rich applications such as geographic information systems (GIS), computer-aided design (CAD), engineering, and graphic design tools. Historically, these workloads necessitated bare-metal systems with their own network connections. However, with virtualization, apps are streamed, and virtual desktop infrastructure (VDI) can assist in the creation of sophisticated desktop experiences.

Security Services-In today's virtualization ecosystem, certain virtual services that run under the network layer are supported by the technology. For SDN platforms to be effective, functions such as NFV must be incorporated. This form of network security enables a truly proactive environment that is capable of minimizing risk and responding to incidents considerably more quickly than is now possible. When a breach happens, every second counts when it comes to preventing the attack from continuing.

10 of 11

Conclusion

SDN has demonstrated a technique to utilize NFV with identical advancements in network and computation capacities to provide a new method of managing modern networking infrastructure. The SDN architecture automatically disconnects the control of data plane transmission from the decision processes associated with distributed applications and policy applications, allowing for greater flexibility in network management. Therefore, SDN and NFV work together so that the ensuing sophistication can be overly effective. With all of this multifunctional software operating the network, the security vulnerabilities and risk are significantly increased, as is the chance of a cyberattack.

11 of 11

References

Haleplidis, E., Pentikousis, K., Denazis, S., Salim, J. H., Meyer, D., & Koufopavlou, O. (2015). Software-defined networking (SDN): Layers and architecture terminology. RFC 7426.
Valdivieso Caraguay, Á. L., Benito Peral, A., Barona Lopez, L. I., & Garcia Villalba, L. J. (2014). SDN: Evolution and opportunities in the development IoT applications. International Journal of Distributed Sensor Networks, 10(5), 735142.
Xie, J., Yu, F. R., Huang, T., Xie, R., Liu, J., Wang, C., & Liu, Y. (2018). A survey of machine learning techniques applied to software defined networking (SDN): Research issues and challenges. IEEE Communications Surveys & Tutorials, 21(1), 393-430.