Improving efficiency and reliability using SDN in multi-layer networks

Abstraction is a big issue in Software Defined Networking (SDN), which is an effective tool to simplify and effectively mask the complexity of network internal properties to achieve fast services.

But for efficient, reliable, and secure network engineering operations, abstraction is useless, but transparency is extremely valuable. Network operators and their supporting systems must discover the full characteristics of the traffic infrastructure in order to analyze and use it.

The true network topology must be discovered in all layers of the physical layer. Multi-layer SDN controllers can solve many problems in network engineering operations by providing this discovery function and using the discovered information.

For example, new service opportunities are emerging in the context of distributed cloud architectures, data center interconnects, and 5G networks, all of which will be supported by IP/optical transport and network services that must be large-scale, delivered quickly, securely, reliably, and with QoS guarantees.

[[214197]]

These and other evolving requirements of IP/optical networks are driving a renaissance in IP/optical integration. This renaissance is primarily seen at the control and management level. This is enabled by a number of developments and advances in network and management technologies, but two stand out as the most important:

Next-generation optical hardware with significant advances in automation and instrument capabilities

The introduction of SDN brings about a developmental, centralized, and distributed control and management software architecture

These technological advances allow us to discover the nature of traffic infrastructure and address and mitigate many of the challenges operators face.

Challenges of IP/Optical Networks

In IP/optical networks, discovering, capturing and accurately visualizing the combined topology and interconnection of the network from Layer 0 to Layer 3 has become a huge challenge for operators. With the passage of time and the development of technology, more complex network architectures have emerged. These complex network architectures have brought serious operational problems, such as unpredictable failures affecting critical network traffic and suboptimal latency control.

Operators need reliable tools to discover, visualize and explore multi-layer topologies and internal connections, which will enable them to better use their network assets. It can help operators overcome the following challenges:

• Inefficient traffic routing: Operators must exercise dynamic control over their network assets so that the network can run more efficiently and services can fully utilize bandwidth. The gap between service provisioning and network engineering must be bridged so that operators can meet dynamic business demands while maximizing the use of available IP/fiber assets.
• Lack of effective protection: In IP/optical networks, a full set of risk scenarios at the optical transport layer (layer 0, layer 1) must be discovered to simplify scenario planning and improve network reliability. For example, a multi-layer protection strategy must be designed to protect traffic at the most cost-effective layer without expensive 1:1 redundancy.
• Cumbersome maintenance coordination: An automated, multi-layered maintenance process is required to minimize or even prevent the negative impact on user traffic during maintenance.
• Sub-maximum latency: Latency information must be collected from the optical network layer to combine the transport layer contribution with the overall latency budget to enable optimization to meet SLA targets.

The benefits of multi-layer SDN

SDN is a revolutionary technology with real value, especially in applications that require extensive visibility and control. In the emerging SDN architecture, multi-layer SDN controllers appear to be optimizing the operation of the optical and IP network layers.

Many optical devices today require embedded operations and management (OAM) protocols to reliably detect topology and connectivity details and communicate with SDN controllers. Link Layer Discovery Protocol (LLDP) is used by network devices to advertise their identity, capabilities, and neighbors.

LLDP snooping is an important discovery tool for data center interconnect applications and WAN architectures. LLDP snooping has many benefits, including avoiding tedious manual configuration and configuration verification, and can also detect misconfigurations. For example, in traditional installations, heuristic techniques can be used to discern the topology and connectivity of network layers where LLDP is not available.

These discovery capabilities can enable solutions to address the above challenges. A multi-layer SDN management architecture can bring a unified view to the management of IP/optical networks.

A centralized, global view provides a clear understanding of how traffic is designed along network paths, which facilitates comprehensive route calculations with the goal of using network resources most efficiently.

A typical rendered architecture shows services on the IP layer and optical network. Detailed information can be obtained from both the IP and optical domains, and new insights into network operations can be synthesized, especially in the area of ​​risk assessment. At the IP layer, path diversity is obvious, but a deeper understanding of the actual optical infrastructure can reveal potential risks, such as when two apparently disjoint optical paths pass through a common fiber, duct, or optical layer forwarding equipment.

[[214198]]

Visualization can be used to ensure minimal impact on IP services during planned maintenance activities, and these activities at the IP layer and optical network layer can be coordinated and orchestrated, such as keeping IP traffic away from affected optical links during actual maintenance.

The ability to detect and accumulate network latency for traffic presents a real competitive advantage, especially as we move into the 5G era. Many 5G use cases, such as AR/VR and remote control systems, rely on consistent and low latency.

SDN brings a range of tools to the automation and optimization of communication networks. Multi-layer discovery within SDN can provide deep visualization of network operations.