The Role of SDN

In contrast to the decentralized control logic that underpins the construction of the Internet as a complex bundle of box-centric protocols and vertically integrated solutions, software-defined networking (SDN) advocates the separation of control logic from hardware and its centralization in software-based controllers. Introducing innovative applications and incorporating automatic and adaptive control into these fundamental tenets can ease network management and enhance user experience.

Challenges to Networking

Over the past few years, there has been a growing demand for a new approach to networking to address the many issues associated with current networks. According to the SDN approach, networking operations can be simplified, network management can be optimized, and innovation and flexibility can be introduced.

According to Kim and Feamster (2013), four key reasons can be identified for the problems encountered in managing existing networks:

1) Complex and low-level network configuration: Network configuration is a distributed task typically configured vendor-specific at the low level. Moreover, network operators constantly change configurations manually due to the rapid growth of the network and changing networking conditions, adding complexity and introducing additional configuration errors to the configuration process.

3) Growing complexity and dynamic network state: networks are becoming increasingly complex and more extensive. Moreover, as mobile computing trends continue to develop and network virtualization (Bari et al. 2013; Alam et al. 2020) and cloud computing (Zhang et al. 2010; Sharkh et al. 2013; Shamshirband et al. 2020) become more prevalent, the networking environment becomes even more dynamic as hosts are constantly moving, arriving and departing due to the flexibility offered by virtual machine migration, which results in a rapid and significant change of traffic patterns and network conditions.

(3) Exposed complexity: today’s large-scale networks are complicated by distributed low-level network configuration interfaces that expose great complexity. Many control and management features are implemented in hardware, which generates this complexity.

(4) Heterogeneous: Current networks contain many heterogeneous network devices, including routers, switches, and middleboxes of various kinds. As a result, network management becomes more complex and inefficient because each appliance has its proprietary configuration tools.

Because legacy networks’ static, inflexible architecture is ill-suited to cope with today’s increasingly dynamic networking trends and meet modern users’ QoE expectations, network management is becoming increasingly challenging. As a result, complex, high-level policies must be adopted to adapt to current networking environments, and network operations must be automated to reduce the tedious work of low-level device configuration.

Traffic Engineering

Networks with multiple Border Gateway Protocol (BGP) Autonomous Systems (ASNs) under the same administrative control implement traffic engineering with policy configurations at border edges. Policies are applied on multiple routers distributedly, which can be hard to manage and scale. Any per-prefix traffic engineering changes may need to occur on various devices and levels.

A new BGP Software-Defined Networking (SDN) solution introduced by P. Lapukhov and E. Nkposong proposes a centralized routing model. It introduces the concept of a BGP SDN controller, also known as an SDN BGP controller with a routing control platform. No protocol extensions or additional protocols are needed to implement the SDN architecture. BGP is employed to push down new routes and peers iBGP with all existing BGP routers.

BGP-only Network

A BGP-only network has many advantages, and this solution promotes a more stable Layer 3-only network, utilizing one control plane protocol – BGP. BGP captures topology discovery and links up/down events. BGP can push different information to different BGP speakers, while an IGP has to flood the same LSA throughout the IGP domain.

For additional pre-information, you may find the following helpful:

1. OpenFlow Protocol
2. What Does SDN Mean
3. BGP Port 179
4. WAN SDN

Back to basics with BGP SDN

BGP Peering Session Overview

A BGP neighbor relationship is called a peer relationship in BGP terminology, unlike OSPF and EIGRP, which implement their transport mechanism. In place of TCP, BGP utilizes BGP TCP port 179 as its transport protocol. A BGP peering session can only be established between two routers after a TCP session has been established between them. Selecting a BGP session consists of establishing a TCP session and exchanging BGP-specific information to establish a BGP peering session.

A TCP session operates on a client/server model. On a specific TCP port number, the server listens for connection attempts. Upon hearing the server’s port number, the client attempts to establish a TCP session. Next, the client sends a TCP synchronization (TCP SYN) message to the listening server to indicate that it is ready to send data.

Upon receiving the client’s request, the server responds with a TCP synchronization acknowledgment (TCP SYN-ACK) message. Finally, the client acknowledges receipt of the SYN-ACK packet by sending a simple TCP acknowledgment (TCP ACK). TCP segments can now be sent from the client to the server. As part of this process, TCP performs a three-way handshake.

So, how does BGP work? BGP is a path-vector protocol that stores routes in the Routing Information Bases (RIBs). The RIB within a BGP speaker consists of three parts:

1. The Adj-RIB-In,
2. The Loc-RIB,
3. The Adj-RIB-Out.

The Adj-RIB-In stores routing information learned from the inbound UPDATE messages advertised by peers to the local router. The routes in the Adj-RIB-In define routes that are available to the path decision process. The Loc-RIB contains routing information the local router selected after applying policy to the routing information in the Adj-RIB-In.

Lab Guide on BGP Route Reflection

The following lab guide will look at the famous BGP RR if you don’t want a full mesh of iBGP speakers.

Route reflectors (RR) are one method to eliminate the full mesh of IBGP peers in your network. The other method is BGP confederations. The route reflector allows all IBGP speakers within your autonomous network to learn about the available routes without introducing loops.

In the example below, we have 3 IBGP routers. With standard IBGP rules, when R2 receives a route from R1, it will not be forwarded to R3 (IBGP split horizon). We will configure R2 as the route reflector to get around this.

Benefits of BGP Route Reflectors:

1. Scalability: Using BGP RRs, network administrators can significantly reduce the number of iBGP sessions required to maintain full connectivity within an AS. This results in a more scalable network architecture, as the complexity of managing and maintaining a full mesh of iBGP connections is eliminated.

2. Reduced Resource Consumption: With BGP RRs, the burden on individual routers to maintain iBGP sessions is alleviated. Instead, the RRs are responsible for reflecting BGP updates to the appropriate routers within the AS. This reduces the processing and memory requirements on the individual routers, freeing up valuable resources.

3. Simplified Configuration: Implementing BGP RRs simplifies the configuration process by centralizing the distribution of BGP updates. Rather than configuring iBGP sessions between every router within the AS, administrators only need to establish iBGP sessions with the RRs. This streamlined configuration process saves time and reduces the potential for misconfigurations.

The Emergence of BGP in SDN:

Software-defined networking (SDN) introduces a paradigm shift in managing and operating networks. Traditionally, network devices such as routers and switches were responsible for handling routing decisions. However, with the advent of SDN, the control plane is decoupled from the data plane, allowing for centralized management and control of the network.

BGP plays a crucial role in the SDN architecture by acting as a control protocol that enables communication between the controller and the network devices. It provides the intelligence and flexibility required for orchestrating network policies and routing decisions in an SDN environment.

Benefits of BGP SDN:

1. Simplified Network Management: BGP SDN simplifies network management by centralizing control and configuration. This allows network administrators to easily define and enforce policies across the entire network, reducing complexity and improving operational efficiency.

2. Scalability and Flexibility: BGP SDN offers enhanced scalability and flexibility compared to traditional networking approaches. With BGP, network administrators can dynamically adapt the routing policies based on network conditions, ensuring optimal traffic flow and load balancing.

3. Improved Network Security: BGP SDN provides enhanced security features by allowing fine-grained control over network access and traffic routing. It enables the implementation of robust security policies, such as traffic isolation and encryption, to protect against potential threats.

4. Increased Network Resilience: BGP SDN improves network resilience by enabling automated failover mechanisms. In a network failure, the centralized controller can efficiently reroute traffic, ensuring uninterrupted connectivity and minimizing downtime.

Layer-2 and Layer-3 Technologies

Traditional forwarding routing protocols and network designs comprise a mix of Layer 2 and 3 technologies. Topologies resemble trees with different aggregation levels, commonly known as access, aggregation, and core. IP routing is deployed at the top layers, while Layer 2 is in the lower tier to support VM mobility and other applications requiring Layer 2 VLANs to communicate.

Fully routed networks are more stable as they confine the Layer 2 broadcast domain to certain areas. Layer 2 is segmented and confined to a single switch, usually used to group ports. Routed designs run Layer 3 to the Top of the Rack (ToR), and VLANs should not span ToR switches. As data centers grow in size, the stability of IP has been preferred over layer 2 protocols.

• A key point: Traffic patterns

Traditional traffic patterns leave the data center, known as north-to-south traffic flow. In this case, conventional tree-like designs are sufficient. Upgrades consist of scale-out mechanisms, such as adding more considerable links or additional line cards. However, today’s applications, such as Hadoop clusters, require much more server-to-server traffic, known as east-to-west traffic flow.

Scaling up traditional tree topologies to match these traffic demands is possible but not an optimum way to run your network. A better choice is to scale your data center horizontally with a CLOS topology ( leaf and spine ), not a tree topology.

Leaf and spine topologies permit equidistant endpoints and horizontal scaling, resulting in a perfect combination for optimum east-to-west traffic patterns. So, what layer 3 protocol do you use for your routing design? An Interior Gateway Protocol (IGP), such as ISIS or OSPF? Or maybe BGP? BGP’s robustness makes it a popular Layer 3 protocol for reducing network complexity.

How BGP works with BGP SDN: Centralized forwarding

What is BGP protocol in networking? Regarding internal data structures, BGP is less complex than a link-state IGP. Instead of forming adjacency maintenance and controls, it runs all its operations over Transmission Control Protocol (TCP) and uses TCP’s robust transport mechanism.

BGP has considerably less flooding overhead than IGPs, with a single flooding domain propagation scope. BGP is great for reducing network complexity and is selected as this SDN solution’s singular control plane mechanism for these reasons.

Peter has written a draft called “Centralized Routing Control in BGP Networks Using Link-State Abstraction,” discussing the use case of BGP for centralized control of routing in the network.

The main benefit of the architecture is centralized control rather than distributed. There is no need to configure policies on multiple devices. All changes are made with an API in the controller.

A link-state map 

The network looks like a collection of BGP ASN, and the entire routing is done with BGP only. First, BGP builds a link-state map of the network in the controller memory.

Then, they use BGP to discover the topology and notice link-up and link-down events. Instead of installing a 5-tuple that can install flows based on the entire IP header, the BGP SDN solution offers destination-based forwarding only. For additional granularity, implement BGP flow spec, RFC 55745, entitled “Dissemination of Flow Specification Rules.” 

Routing Control Platform

The proposed method was inspired by the Routing Control Platform (RCP). The RCP platform uses a controller-based function and selects BGP routes for the routers in an AS using a complete view of the available routes and IGP topology. The RCP platform has properties similar to those of the BGP SDN solution.

Both run iBGP peers to all routers in the network and influence the default topology by changing the controller and pushing down new routes. However, a significant difference is that the RCP has additional IGP peerings. It’s not a BGP-only network. BGP SDN promotes a single control plane of BGP without any IGPs.

BGP is used to detect health, build a link-state map, and represent the network to a third-party application as multiple topologies. You can map prefixes to different topologies and change link costs from the API.

Multi-Topology view

The agent builds the link-state database and presents a multi-topology view of this data to the client applications. You may clone this topology and give certain links higher costs, mapping some prefixes to this new non-default topology. The controller pushes new routes down with BGP.

The peering is based on iBGP, so new routes are set with a better Local Preference, enabling them to be selected higher in the BGP path decision process. It is possible to do this with eBGP, but iBGP can be more accessible. With iBGP, you don’t need to care about the next hops.

BGP and OpenFlow

What is OpenFlow? BGP works like OpenFlow and pushes down the forwarding information. It populates routes in the forwarding table. Instead of using BGP in a distributed fashion, they centralize it. One main benefit of using BGP over OpenFlow is that you can shut the controller down, and regular BGP operation continues on the network.

But if you transition to an OpenFlow configuration, you cannot roll back as quickly as you could with BGP. Using BGP inband has great operational benefits. It is a great design by P. Lapukhov. There is no need to deploy BGP-LS or any other enhancements to BGP.

Future Outlook: As the demand for more agile and efficient networks continues to grow, BGP SDN is expected to play a pivotal role in shaping the future of networking. Its ability to simplify network management, enhance scalability, and improve security makes it an ideal choice for organizations seeking to modernize their network infrastructure.

BGP SDN represents a significant advancement in networking technology, allowing organizations to build agile, scalable, and secure networks. By centralizing control and leveraging the intelligence of BGP, SDN has the potential to revolutionize how networks are managed and operated. As the industry embraces SDN, BGP will continue to play a crucial role in enabling the next generation of network infrastructure.