Benefits of Not Being Completely Vertical
Additional thoughts coming out of NGON & DCI Europe
By the title you might infer that I am referring to relaxing on the beautiful beaches of Nice. While there are certainly benefits to that component of not being vertical, in this case vertical refers to vertical integration whereby one supplier creates everything in their ecosystem – from components to equipment.
In one of Light Reading’s video recaps of NGON 2018, they gave a nice shout out to my comments about the future of optics in 5G access networks (it’s at the 3:34 mark is this video). I thought it might be useful to provide some more context for those who were not in the room for the 5G breakfast panel discussion on Tuesday morning. It was an interesting topic that sparked a vigorous discussion both in the room and later on the show floor.
The question asked of the panel was specifically about lower cost coherent optics (aka “coherent lite”) for use in locations where the distance requirements are not as stringent but the cost requirements are. The topic came up in a discussion of optics for 5G transport networks, where enormous amounts of bandwidth are going to be needed in the access part of the network. Whereas there are a variety of 100Gbps and faster optics choices currently on the market or about to be introduced, they generally fall into two categories – coherent which works over longer distances and requires only one wavelength, and direct detect which is used mostly inside a building/CO/data center and can require multiple wavelengths (this is a slight simplification for a blog post, but generally correct). Coherent is expensive while direct detect is less so. When considering 100Gbps and higher rates for 5G backhaul the preference is for something with a single wavelength that can travel over a longer distance (hence coherent), but the cost requirements are especially tough to meet (hence direct detect).
The panel and audience had been discussing the need for a lower cost coherent optical technology and the question was asked if anyone saw a place for direct detect in the 5G transport market. That question prompted my answer, “we (ECI) really don’t care, and we’ll use whatever is the lowest cost with the required performance.” The technology inside the optics really does not matter. It is the performance and the cost that matter, and we will take the component that best meets those requirements. This is a huge benefit of being a company that is not reliant on our own internal optics development for our success. We – and others like us – are able to pick and choose the best of breed components for any aspect of our customer’s networks.
To be clear, we love our optics suppliers. We work closely with them to ensure that their roadmaps match our roadmaps and that we are getting the best performance and cost for our customers. However, that doesn’t mean that we are tied exclusively to a single technology for every application. For customers who want long-distance and high performance operation, we believe that we have the best possible coherent optical technology for that application. If at some point a bright optical engineer in the research department of a component supplier comes up with a way to make a lower cost, lower performance single wavelength solution that meets our customers’ needs, we will be glad to work with them to implement that technology into our platforms.
There are certainly benefits to being vertically integrated – otherwise no one would do it. Being vertically integrated means the ability to co-develop technologies and the component and systems levels, early access (potentially) to new technology, theoretically higher margins, and no dependence on someone else’s development priorities. For companies that live or die by the few month advantage (admittedly occasionally more) that vertical integration gives, their decision makes sense. For others the decision to choose from a best-of-breed solution allows greater flexibility, especially in a dynamic market like 5G.
What technology will finally win out in the complex and rapidly changing world of 5G transport networks is still very much to be decided. The requirements are difficult to meet and many operators are considering very different approaches. What is very clear is that the winning technology will not win just because of the acronyms associated with the transistors and lasers inside the module. The winning technology will win because it meets the needs of the network operators at a price point that they can afford. Having the flexibility to choose among the options that will be introduced over the next few years is a huge advantage.