How to Design a Highly Reliable Fiber Optic Network
What happens if a major fiber optic cable is cut or a major hubbing location is destroyed
in a fiber network? Will the whole system be brought down? That is the subject of this
article: the survivability of a well designed fiber network.
Modern fiber optic transmission systems have provided tremendous capacity for voice,
video and data traffic. At the same time, the fiber network needs to be designed to take the
full advantage of this economies of scale while minimizes maintenance cost.
The main design goal is to utilize cost-effective interoffice (facilities between buildings)
to maximize the number of circuits remaining intact after a worst-case failure while minimizing
the ongoing maintenance cost associated with this survivability.
Conventional (existing) network architecture and design methods as well as future technologies
can both be used to achieve this goal to some different degrees.
The principles in designing a survivable fiber network
Put simply, the basic principle behind the design of efficient fiber networks is "demand
aggregation". Demand aggregation makes use of facility hubbing and facility hierarchies to
optimize facility networks.
What is facility hubbing?
Facility hubbing depends on the fact that fiber optic equipment occupies the majority
cost in a fiber network. A fiber optic network's total cost is not very sensitive to the network's
distance. In the mean time, fiber's capacity is too much for building-to-building data traffic.
As a result, it is concluded that routing traffic from each building into a building selected
as its "hub" is the most efficient network design. In this way, traffic is aggregated into the
largest possible bundle to take advantage of economies of scale of fiber technology.
At the hub, all traffic is sorted and properly assigned to get to its destination. Digital
Cross-Connect (DCS) is used at the hub to rearrange lower rate traffic into higher rate traffic.
Thus all traffic is concentrated into high capacity routes to a central location where the
demands are sorted according to destination.
What is facility hierarchy?
Facilty hierarchy is the extension of facility hubbing. The concept of facility hubbing
is brought a step further. In this concept, buildings are grouped into "clusters", with each
cluster having one hub building. This approach considers such factors as community of interest
and geographic area.
The clusters can then be regrouped into "sectors", with each sector having a "gateway".
The gateway is actually a hub building designated to handle inter-cluster traffic demand. A
gateway can then aggregate demand from several cluster hubs to form a large demand to be routed
to another gateway in much the same way as demand is aggregated to a hub building.
Actually the concept of facility hierarchy can be extended to an arbitrary number of levels.
Current fiber optic technology trends have forced a radical change in the structure of
networks and the way in which networks are designed. The large capacity of fiber optic systems
have resulted in efficient networks having large aggregations of demand in a few facilities.
Properly designed network architecture can increase network cost slightly but increase
network survivability greatly.
Colin Yao is an expert on fiber optic communication technologies and products. Learn
optic breakout kits on Fiber Optics For Sale Co. web site.
More Network Troubleshooting and Support Articles:
• Standard Network Path Metrics
• What is the Difference Between NAT and PAT?
• Five Open Source DevOps Tools
• Wireless Network Troubleshooting
• Five Network Design Considerations
• Turning Names (URLs) Into IP Addresses
• Configuring DHCP for Your SME Network
• Remote Connectivity
• Putting Your SME Data on the Internet
• Local Network IP Addressing Strategies