Getting the Most Out of Your Network: Network Design Considerations

Dr. DaveBroadband

ethernet cable connected to a swtich

For those of you who know me, I am maniacal about technology and seeking out ways in which it can help both business and society.  A well-designed data network can make a world of difference when it comes to communication and productivity.  Here are a few design considerations that I often discuss with customers.  While not a comprehensive list, these are important considerations at the start of any network build process and will ensure your network is best configured to meet your organization’s needs both now and into the future.

1. Start with the Big Picture

Like with any project, it pays to start with the big picture. Instead of designing a network that meets only your current needs, look for ways you and other potential users could leverage your network in the future. For example, when mapping out the physical path for your network, consider potential users (such as police and fire stations or libraries) and consider placing handholes nearby.

By judiciously selecting which streets to use and where to turn a corner, you can dramatically increase your potential revenue without having to shoulder additional costs.

2. Invest in Your Infrastructure

As the old saying goes, you have to spend money to make money. Of course not all spends produce the same return on investment.  By investing appropriately in your new broadband network you may be able to increase your revenues both now and in the future. As mentioned above, starting with the big picture (instead of building out your network piecemeal), you can increase efficiency, gain more revenue and save money in the long run. Here are some specific suggestions:

a) Small Fibers; Big Impact

For any fiber network design one should always consider using conduit with microducts.  The flexibility to add (and remove) fiber as your needs grow often out weighs the nominal cost increase.  Additionally, one should consider the pro’s and con’s of filling the above microducts with micro-cable fibers.  These are loose tube fibers with reduced fiber diameter (200-µm, compared to 250-µm) and reduced buffer tube diameter (1.4 mm compared to 2.5 mm). 

Since the cross-sectional area is proportional to the square of the diameter you can fill the same microduct with more than twice as many micro-cable fibers compared to standard loose tube fibers. This increase allows you to install either more total fibers or use a smaller diameter conduit for the same number of fibers. Microduct routes can often be installed in already occupied ducts, making it easier to replace old cables in your network without disrupting service.

They not only allow you to replace cables as necessary (either because your current cables have become damaged or you are upgrading your network from copper cable to fiber optic cables) they also make it easier to grow your network on demand.

b) The Benefits of Second Conduits

When designing your network, you may also want to consider placing a second conduit with microtubes for future leasing or horse-trading. This could, in principle, pay for itself under the right circumstances. The general idea is that you bury two conduits side-by-side but you build only one of them into a working network (i.e. w/ handholes, fiber, etc.) and reserve it for your exclusive use.  You can then offer to other customers, partners or competitors the use of microducts in the 2nd conduit.  As an example of horse trading  you could give the first customer a reduced price if he installs all the handholes, fiber, etc. for the 2nd conduit.  Revenue from additional customers can further defray your initial investment costs.

3. Build Redundant Networks (if at all Possible)

Redundant networks ensure that service stays up or that downtime is minimized if a critical link or component fails. While redundancies may seem like costly or unnecessary additions to your network, they can provide additional revenue.  Increasingly, customers value this reliability and are willing to pay a premium for it. 

In spite of these benefits the additional funds that network redundancy requires is often hard to come by.  Here are three creative ways to get around this problem:

a) Leverage All Your Options to Build Redundancy

In addition to your traditional customers who pay monthly service fees, other users can also benefit from your network build and be willing to pay for that benefit. For example, Municipalities, emergency service entities or even other service providers may be willing to pay for some of your construction costs in exchange for access to your new microducts.  Our engineers often give advice as to who might be interested and what they may be willing to pay.

b) Consider Fixed Wireless

Fixed wireless can be a great option for a potential redundant path or to complete the last leg of a wired redundant path.  The reach of a given wireless hop depends on a number of factors.  What often drives the design of a wireless segment is the required bandwidth.  As a rule of thumb, you can expect 10 Gbps over a 2-mile link or 1 Gbps over a 10-mile link. 

Even a lower-bandwidth/longer-reach link that maintains connectivity (albeit at a lower rate) can keep many of your customers satisfied during an outage.  For these reasons, wireless options should be considered for augmenting or enhancing a fiber network build.

c) The Benefits of Ethernet Ring Protection Switching

Any Ethernet network that has redundancy requires some protocol that prevents loops.  Ethernet assumes there is only one way to get from any point A to any point B.  Any violation of this assumption, like a loop, quickly brings down the Ethernet network.   There are several options for ensuring there are no loops but we recommend Ethernet Ring Protection Switching (EPPS) also known by its ITU-T designation G.8032.  Briefly, ERPS prevents loops by designating one of the links on the ring as the Ring Protection Link (RPL) and blocks traffic on that link under normal conditions.  As long as this link is blocked there can be no loop.  Protection switching is achieved by monitoring the other links (using various tricks) for signs of a failure.  In the event of a link failure ERPS unblocks the RPL link and blocks the failed link.  As a result there is always one link that is blocked and hence there is never a loop.

Though it may sound complicated, this protocol is relatively simple to install, available on most commercial switches and is widely used throughout the telecom industry.

In summary, a well-planned network with appropriate redundancies in place is worth its weight in gold. Planning ahead can reduce or eliminate potential problems later on and help you better attract and serve future customers. This increases your revenue and allows your customers to cover your initial setup costs as well as any upgrades, repairs, or maintenance costs later on.