Top 5 Trends in Optical Networking
In this article
Considering an optical transport system today can be a daunting task. With so many options available like open line systems, compact modular and the convergence of optical and packet, it can be overwhelming, to say the least. What should a network operator consider when thinking about a next-generation optical transport system? This article will break down our top five things you need to know about a next-generation optical transport system.
The old saying less is more definitely applies when it comes to compact modular. For the past several years, there has been a boom in compact modular platforms and sales. It seems every vendor has its own variety and flavor, be it the Cisco NCS1000, Infinera Groove or Nokia 1830 PSI, to mention a few. To match this boom in the product development of compact modular platforms, sales have also skyrocketed.
But why the move towards compact modular? What isn't working that was working in the past? For starters, space, power and cooling. Large bulky systems of the past required massive rack space, power, and cooling to operate. Moreover, regardless of how little functionality a site required, a systems operator was forced to implement a large system. These archaic strategies were pushed down to OEMs by the largest operators, which required extensive systems due to their enormous scale of operations. As the industry found out, this strategy does not work for everyone. Thus equipment vendors started to develop smaller-scale compact modular systems that can address any operator's requirements, no matter their size. With compact modular systems, operators can pick and choose each "building block" for the right situation. If you need a large capacity Muxponder? Easy! How about an add/drop with 48 channels of DWDM? No problem.
In some situations, this may require a single pizza box (1RU) modular chassis. In other cases, it may take multiple chassis to satisfy the needs of one location. But the idea is simple, almost a la carte, if you will. Just pick from the menu of optical transport appetizers until you're satisfied.
Open line systems (OLS) and disaggregation allow the operator to choose the best, most relevant solutions for each portion of the network they are addressing. As seen in Figure 2, regardless of the block of the network an operator is looking to address, we can leverage open APIs to have a single network management system (NMS) to provision and monitor the system and services.
This enables the network operator to deploy multi-vendor hardware in the same network with seamless interoperability. It is widely known that vendor technologies' cycles, direction, research and development, and survivability can be wildly different. With this flexibility, network operators are no longer "locked" into one specific vendor's technology. If you deployed equipment from vendor "a" but a year later vendor "b's" product is a better fit, you can easily integrate vendor "b" into your existing network.
A single vendor can also be acquired by another entity or simply go belly up. In situations like this, it is never in the best interest of an organization or its shareholders to stick to a single vendor solution. Through open line systems and disaggregation, the operator now has the power to choose which solutions work best for them and can quickly shift when needed.
For further information on an open construct, check out my previous article on OLS and disaggregation.
With bandwidth, enough never seems to be enough. It isn't surprising that the bandwidth explosion seems to never stop. 1G was the new 100 Mb, 10G was the new 1G, 100G was the new 10G — so on and so forth until the end of time. Today, however, it would seem that the industry is taking a somewhat brief stop at 400G.
It seems that the form factor size, power, heat dissipation and spectral efficiency very much fits the bill for what the industry is currently looking for. With 400G, an operator can do so much. It can break out into multiple 100Gs and can be our DWDM carrier in routers, switches and optical transport systems. Most importantly, it has the optical performance of 200G while carrying double the bandwidth. All these reasons are why 400G technology makes the most practical sense and why the industry is so focused on developing it. It fits perfectly with a compact, modular, open, flexible, spectral efficient and high bandwidth solution operators are craving.
We don't doubt that 800G will be just around the corner with the same performance, space and efficiency as 400G currently has. In the meantime, 400G is a perfect fit and will easily integrate with 800G because they are both being developed based on the same technology.
When one considers flexibility, a Swiss army knife comes to mind. Operators are in desperate need of a modular solution capable of many different types of applications over various portions of their network. Networks are no longer entertaining a one-size-fits-all mentality but rather looking for an a la carte menu option. The operator can deploy a single solution, compact and modular, while also supporting a vast range of functions. In optical transport systems, there is a dynamic range of functions across the system such as DWDM, ROADM or OTDR to name a few. These functions are mostly dependent on the number of services being deployed at any given location or across the network and distances between nodes. So given an a la carte option, an operator can scale the location to the appropriate size.
For example, below, we have two data centers located across a relatively long distance. The network operator needs to transport 9.6 Tb between them to support mainframe backup, content sharing and greater network resiliency. Given the flexibility of a compact modular design, the operator can stack a series of muxponders feeding into a DWDM filter, and then add amplification to address the long distance between the data centers.
The flexibility of a compact modular solution isn't limited to the above example. An operator can take advantage of a variety of options to deploy in their solutions:
- Open Line System
- Dense Wavelength Division Multiplexing (DWDM)
- Raman Amplification
- Optical Service Channel (OSC)
- Optical Channel Power Monitoring (OCPM)
- Optical Time Domain Reflectometer (OTDR)
- Wavelength Selective Switch (WSS) or Reconfigurable Optical Add/Drop Multiplexer (ROADM)
- Contentionless, Directionless, Colorless ROADM (CDC)
- Flex Spectrum
Since the beginning of optical transport systems, DWDM muxponders and transponders have always reigned supreme — given their ability to extend the reach of 1/10/100G and beyond interfaces over longer distances while also adding an ITU grid DWDM wavelength to that signal. Simply put, routers did not have the technology available to compact these large and bulky functions and shrink them down into a pluggable interface housed directly into the router itself with the same extended reach and FEC capabilities over a DWDM wavelength. That is, of course, until now.
Today's technology OEMs like Cisco can utilize 100/200/400G digital coherent optics (DCO) or 400G QSFP-DD ZR/ZR+ optics and stick them directly into routers. With this new capability, routers can now simply bypass the transponder or muxponder of a traditional optical transport system. However, this is not the only benefit of using such a strategy. Routers also have the inherent benefit of utilizing multiprotocol label switching (MPLS) or segment routing (SR-MPLS). Simply put, MPLS or SR-MPLS gives the router the ability to switch signals through routers creating a path from source to destination.
With ROADM, the signals would optically be passed through nodes. However, with routers, it would electronically be switched. Different strategies, same result. A client service (1/10/40/100/400G) is given a path from source to destination. This, in a nutshell, is the convergence of optical and packet.
With this strategy, we could, in theory, simply replace the functionality of a ROADM system entirely. If our routed layer could transport high bandwidth DWDM interfaces while allowing sub-rate client interfaces to be switched through routers electronically, there is almost no need for ROADMs. They are for all intents and purposes achieving the same goal.
However, merely housing DCO/ZR/ZR+ optics into routers doesn't solve all our transport issues just yet. There is still the issue of multiplexing and amplifying these optics across longer distances. Pluggable DCO/ZR/ZR+ can, in some cases, have a reach of up to 80 kilometers (50 miles). However, these optics also cannot simply multiplex themselves either. To assist in solving this dilemma, the router can be accompanied by external multiplexing filters (i.e., 40/64/80 channel filters) to allow for multiple high bandwidth DWDM interfaces. These filters can be as small as 1 or 2 rack units and are typically passive non-powered devices. This solves the scale issue. With external filters, we would allow for many DWDM wavelengths to be multiplexed in between router locations to scale the capacity of our outside plant fiber pair.
Now comes the issue of extended-reach applications. This is where two routers are connected over a longer distance of outside plant fiber. For this, an external amplifier can be placed. An amplifier can simply boost up the composite signal of all our DWDM interfaces to achieve end-to-end distance limitations.
All in all, a converged optical + packet solution would, at its core, house digital coherent optics (DCO) or ZR/ZR+ 400G optics in a router. For applications where we would want to expand the fiber capacity, an external multiplexer can be placed to combine many wavelengths. Finally, if we have extended distance applications where end to end router optics will not be able to reach, an external amplifier can be added to the solution to achieve full end-to-end connections while leveraging SR/MPLS to switch client sub-rate traffic to and from their endpoints throughout the network infrastructure.
For further information on converged optical + packet networks and strategy, please see this article on hop-by-hop written by my co-workers John Chiarini and Mike DiVincenzo.
Here at WWT, we are keeping an ear to the ground to see what OEMs are doing to keep up in the next-generation optical transport system race. It's an exciting time to see so much innovation and rapid development taking place. In general, the market is painting a picture, and the OEMs are responding in kind.
We have seen rapid development from the likes of Cisco, Infinera and Nokia to provide solutions that are compact modular, open, highly scalable, flexible and converged. With all of these features in mind, a network operator can decrease the overall footprint, power and cost while increasing the flexibility and programmability of their solution.
We hope this article has given you a better understanding of some things to consider while entertaining a next-generation optical transport system. If you have any questions, please feel free to contact us.