The pace of fiber-optic network deployment is accelerating as service providers race to adapt to the growing bandwidth demands of next-generation data transmission. The demand is broad and comes from several sources, including data centers, enterprise and metropolitan area network backbone, as well as powering services like 5G.
With no sign of a slowdown, now is the time to look ahead and consider what fibre networks will need in the next 20 years. From a technical and operational point of view, it is important to determine how much and what amount of fiber is needed for any particular network. While determining the level of capacity required for a particular route is sometimes done in the field, high-quality networks are planned based on the types of customers that will be served and the bandwidth they may need. To build a scalable high-capacity network is to plan end-users' needs today while anticipating customers' escalating needs over the next 20 years.
So what is the industry's answer to this growing demand?
Rapid upgrading and overbuilding
The growth of high-bandwidth applications is enabling fiber optic service providers to deploy rapidly. Today's demand planning, while anticipating future demand, requires increasing the number of optical fibers and establishing pathways that can easily accommodate excess construction and upgrades.
And it's not just the number of fibers that need to be upgraded. Today's technology changes so quickly that a new generation of hardware is introduced every few months. As low-latency requirements for new services such as 5G increase, next-generation routing and switching equipment must also keep up with the latest service-level requirements.
In today's age of (super) connectivity, the status quo is a moving target. Instead, it is best to plan your network for situations where you need more throughput than you initially expected. That means overbuilding the network for those services now, or designing the network and preparing paths so that it's easier to deploy more fiber later.
The number of fibers is important
Fifteen years ago, the backbone of the network was 96 optical fibers. Today's network standards call for optical fibers with 288 to 864 cores or more. For example, for data centers with increasing bandwidth requirements, a 3456 core fiber might be used.
When considering the overall cost of construction, optical fiber itself is a small percentage of the total construction, so it makes business sense to overbuild to ensure future capacity. Digging up roads or sidewalks to install more fiber is much more expensive and disruptive than running more fiber in the initial construction.
The key is to strike the right balance. There is a case for overbuilding. For example, if 48 fibers are needed to serve a region today, installing 1,728 fibers may be overkill. Many fiber service providers have adopted a formulaic approach to new construction without committing to massive overbuilding.
Changes in optical fiber
Over the past few years, changes in fiber optics have made it more efficient and can continue to meet future needs. Although the purity of glass has always been static and G652 is still the standard, the physical dimensions of the fibers have changed due to new technologies and processes that reduce their internal mass. Newer fibers do away with the strength components within the fibers and "glue" the strands together, reducing the size of the bundle by about half. For example, a tube that used to be an inch wide is now half an inch and contains the same or more fibers.
This is a huge step forward for the industry. See 432 or 864 fibers using this denser composition has become mainstream. This is particularly true in areas where the Ministry of Transport has a critical crossing point or leases third-party systems that require capacity but have limited space.
Planning for next generation services
Designing scalable networks to meet quality-of-service (QoS) standards requires more than just more fibre optics. Fiber optic service providers are able to manage the frequency of upgrades and maintenance with less interference to end customers using clear metrics around the capacity requirements for a particular situation (e.g. horizontal to single tenant, ring or backbone). The result is consistent and reliable uptime each month.
For fiber service providers, planning for the next 20 years means building sufficient front-end capacity and creating routes that are strategically designed, built and licensed to support their longevity and ability to easily upgrade. This creates continuous QoS that helps customers succeed and scale faster. Eventually, if a service provider builds its network correctly, it will fade into the background. After all, an always-on web means a job well done.