Which Optical Transport Architecture Is Right for You?

There are various transport network architectures, from strictly Layer 1 optical rings to meshed Layer 1 through Layer 3 converged networks. How do you decide? This article provides a high level overview of three different architectures, their typical use cases and benefits.

March 26, 2021 6 minute read

Our last three articles focused on the evolution of optical networking, high-speed pluggable DWDM optics and focused on the hop-by-hop or Routed Optical Network (RON) architecture. I referenced other architectures like ROADM, Rings and Point to Point, so perhaps you are wondering, are those architectures still relevant, and if so, which one is right for my environment? The short answer is yes, they are, and it depends. This article provides an overview of the three most common optical network architectures and their typical use cases. 

1. Reconfigurable optical add-drop multiplexer (ROADM) networks

ROADM networks are the most common type of optical network deployed in the field today. They exist in many forms, including rings, meshes and multi-site point to point. They are chassis-based, self-contained systems that include transponders, ROADM cards and amplifiers. This type of architecture provides the highest level of flexibility, such as service type and size, moving/adding/deleting services and path redundancy. They can also span very long distances, with amplifiers located along the fiber path; ROADM sites can be 1000s of kilometers apart. 

A key benefit of a line card-based system is the variety of line cards available and the ability to mix and match line cards as needed. You can deliver a range of speeds and data types within one chassis, a critical requirement in environments with a wide range of applications. For example, you may need to offer varying GbE rates ranging from 1GbE to 100GbE, along with Fiber Channel for a SAN (storage array network) environment. You can even incorporate Layer 2 capable line cards and provide multi-rate, multi-site L2 VPN services.

ROADM example
Figure 1: ROADM example

ROADM systems also provide the highest wavelength flexibility and efficiency because they enable each site to select wavelengths to carry traffic or permit wavelengths that do not terminate at that site to pass through to its destination. You can remotely add, move or change services on-demand, saving a tremendous amount of time and operational expense. ROADM systems also enable "multi-degree." 

Multi-degree allows you to have multiple fiber paths per node connecting to other sites. Multi-degree provides the ability to "steer" traffic over specific paths and provide dynamic path restoration if a fiber path in the network fails. For example, each site in Figure 1 above has three degrees, forming a mesh between the four sites. In this example, each node has three possible paths to the other sites — one path would be the primary path, with two available restoration paths if the primary path were to fail.

So if you or your customer are looking for a self-contained, highly flexible, fully redundant system with the capability to transport a wide range of data types, a ROADM system is most likely the way to go.

2. Point-to-point networks

A point-to-point DWDM network is a simplified version of the ROADM system. It was born from the need for relatively inexpensive, high-capacity transport between large data centers. Point to point networks utilize a multiplexer/demultiplexer at each site with one pair of fibers connecting the sites. The multiplexer takes in (adds) the individual transponders wavelengths, combines them and transmits them to the other site on one fiber. 

On the other fiber, the demultiplexer receives wavelengths from the other site and separates (drops) them back out. Working together, this provides up to 96+ bidirectional, individual connections or circuits between the two sites. That's the equivalent of 192 individual fibers. This application is widely used for data center-to-data center connectivity. 

Point-to-point example
Figure 2: Point-to-point example

The transponder is the device that takes the electrical Ethernet data stream and converts it to an optical signal at a specific wavelength. There are two transponder options, an independent device like the Cisco NCS-1000 series or a pluggable optic that sits in a router or switch (Figure 2 illustrates both of these options). An independent transponder provides high density 10G, 100G and 400G Ethernet connections between your customer's devices utilizing one or more optical wavelengths. An independent transponder is typically used when you do not own or manage the routers, switches or servers connected across the point-to-point link. They are also used when very high density and multi-rate services are required.

If you own or manage the devices connected by your point-to-point optical system, then pluggable DWDM optics are well worth considering. They eliminate the need to purchase and manage an independent transponder because that function now occurs within the pluggable optic seated in the router or switch. There are 100G, 200G and 400G DWDM data rates available per pluggable optic, which provides flexibility for initial and future bandwidth requirements. Learn more about pluggable DWDM optics. 

If your objective is to transport relatively large amounts of data between two locations at the lowest possible cost per bit, then a point-to-point network is probably the right choice for you. 

3. Routed optical networks (RON)

The routed optical network (RON), also known as "hop-by-hop," is an emerging architecture that combines the point-to-point optical network's simplicity with the port density and massive processing power of today's core routers. This architecture leverages the point-to-point network's ability to provide large amounts of bandwidth between adjacent sites then leverages the router's layer 3 technology and features to provide similar functionality as a ROADM system. Converging layer 1 with layer 3 simplifies the transport network resulting in equipment cost and operational cost savings.

RON example
Figure 3: RON example

Figure 3 above is an example of the RON architecture. The four sites are connected with a point-to-point optical system between adjacent routers, forming a ring. We could easily create a mesh, like the ROADM example, by adding additional point-to-point systems. The DWDM optics are typically seated in the router, making the architecture more cost-effective than deploying an independent transponder. DWDM pluggable optics range from 100G to 400G, making it easy to start small and "pay as you grow." If you need to increase site-to-site bandwidth, you simply install another DWDM optic.

The RON architecture, generally speaking, is best suited for networks that have a relatively even distribution of nodes and bandwidth requirements. Complex meshed networks tend to consume bandwidth disproportionally, resulting in path utilization issues and even prove to be less cost-effective than deploying a ROADM system. 

With that in mind, we are starting to see equipment vendors exploring a hybrid model that incorporates simplified ROADM functionality into the point-to-point architecture, which would allow for more efficient use of router capacity. Get a more in-depth look into the RON architecture. 


There is no cookie-cutter solution for every network environment. Every network has its unique conditions and requirements. That said, it's likely your specific transport network requirements fall within one or a combination of these architectures. If you or your customer are considering building, expanding or refreshing a transport network, WWT can help. We have a solution architects team focused on transport network design and technology who can work with you from network design and vendor selection to deployment and training. Reach out to your WWT account team for more information, or feel free to contact us to learn more.

Share this