Wireline Vs. Wireless:

Wireline vs. Wireless: Exploring Total Cost of Ownership in Outdoor Applications

A Farpoint Group White Paper

Document FPG 2009-202.1 June 2009

Farpoint Group White Paper – June 2009

D

espite the overall macroeconomic picture preoccupying the attention of just about everyone these days, there is no recession in networking. Networks of all forms based on the Internet Protocol (IP), after all, constitute the fundamental circulatory system of the world’s economy. Enterprise and municipal local-area networks (LANs), metropolitan-area networks (MANs), and wide-area data services (WANs) are essential to current operations - and future growth. The American Recovery and Reinvestment Act of 2009 promises billions in new investment for broadband infrastructure, much of this targeted at underserved areas of the United States. Similar initiatives are underway in other parts of the world. Even without such stimulus, however, 2009 would be growth year for networks - they’re that important, and globally. And while many still think of wireless in terms of wireless LANs (with limited range and primary application in the residence and the enterprise) and cellular data (with limited throughput but significant coverage, especially in major metropolitan areas), outdoor broadband systems are today and have been for some time provisioning significant capacity and throughput with remarkable convenience and cost-effectiveness – especially when compared to the wireline alternative. In fact, wireless is often the only solution in metro-area broadband applications simply because wireline services are not available or are the more expensive alternative. As we’ll see in this White Paper, the total cost of ownership of an outdoor wireless network can be remarkably low, and the technology can enable broadband services in locales that otherwise would have remain entirely unserved. Many today thus believe that wireless networks will quite literally dominate much of the broadband access (end-users at the edge of the network) and a significant share of the distribution (elements between the network core and the edge) markets, and we’re certainly inclined to agree. While wire will always play a role in legacy implementations and some interconnect (often called backhaul) and stationary applications, wireless is increasingly becoming the solution of choice, and for a number of reasons, as follows: •

Eliminating restrictions on installation - wireless is the ideal solution in any situation where wire is difficult or impossible to install either physically or economically. This includes not only the installation of networks in structures that are not readily amenable to wiring, but also situations where networks must be rapidly deployed, are subject to ongoing modifications, changes, and upgrades, and where post-installation flexibility is otherwise required. And also included here are outdoor installations where provisioning wire, cable, or fiber involves hanging these on poles, digging up streets, or related expensive, time-consuming activities. Time-to-solution is universally better in the case of wireless.

•

Fixed and mobile access - Access is increasingly being defined in terms of mobility, both indoors and out. Convenience, to say nothing of productivity, depends upon the “anytime/anywhere” capabilities essential in wireless and simply not available via wire. As proof, just consider how the cell phone is becoming the primary, default, or even only access for many people today, and for both voice and data. Wire is thus being marginalized, to a great degree, to

Wireline vs. Wireless: Exploring TCO in Outdoor Applications

1

Farpoint Group White Paper – June 2009 interconnect and applications of a stationary nature, such as call centers, help desks, accounting departments, printers (although wireless is, in fact, making big inroads here) and, of course, inside data centers. Outdoors, we’re seeing progress in fiber (and other cabled broadband) to the home (and of course businesses) in some parts of the world, but this is only cost-effective in areas of dense population. This leaves a huge opportunity for wireless in situations where wire cannot be installed due to physical restrictions and/or simply the cost involved. •

Closing the performance gap - For many years, wireless could not compete with the throughput and quality-of-service capabilities of wire. This has now changed, and radically – for example, wireless LANs based on the 802.11n standard in both indoor and outdoor applications offer much higher throughput than 100 Mbps Ethernet. Outdoors, wireless links capable of bridging several kilometers can yield throughput well in excess of 1 Gbps, and today have the management capabilities required to effectively plan, configure, install, monitor, and report on all elements of an operational broadband wireless solution. All of this means that wireless systems are highly competitive with wired solutions based on xDSL, cable modems, optical fiber, and metro-area Ethernet services.

•

Support for any application - Early wireless links were lucky to yield a few kilobits per second. But today’s high level of throughput available, coupled with support for time-bounded communications like voice telephony and real-time (and even two-way) video based on IP (often called IPTV), mean that wireless solutions effectively duplicate the capabilities of wire, and often, as we’ll see below, exceed them on a price/performance basis. In other words, an appropriate wireless solution can support any application today, and at a price that is more than competitive (and often very significantly better) than that of wire. An application with increasing importance is the provisioning of video for lawenforcement and surveillance applications – no longer a challenge to wireless solutions. But “last mile” access applications using a fixed infrastructure to fixed (and sometime mobile) clients is the classic case for outdoor wireless.

•

Cost-effective deployments and ongoing operations - Finally, and this leads us to the core objective of this White Paper, wireless can be far more cost-effective than wire in outdoor applications. In order to fully appreciate the cost issue, it’s important to consider two key elements in the total cost of ownership (TCO) picture: capital expense (CapEx) and operating expense (OpEx). We’ll cover these in more detail below, but for now consider that all networking installations will have either capital-equipment costs (which may include some changes, upgrades, and additions over time) or service-provider costs (which may include some capital equipment), and usually a combination of these two. Finding the most effective balance between these two elements is the key to minimizing TCO.

In short, wireless technologies and products have reached the point where the undeniable convenience and flexibility inherent in the technology itself make wireless the first choice of so many users and applications. But wireless still has the reputation, unfortunately, of Wireline vs. Wireless: Exploring TCO in Outdoor Applications

2

Farpoint Group White Paper – June 2009 being the more expensive option when compared with equivalent wired services. As we will demonstrate below, such is no longer the case. Wireless can indeed provide the convenience and performance required in essentially any application, and at a cost that is often a significant improvement over that of wire. Evaluating Total Cost of Ownership (TCO) As we noted above, wireless by historical reputation has been assumed to yield much less performance than wire, with the additional assumption of much higher cost in many cases. This situation, however, has changed remarkably in the past few years with the continuing high rate of innovation in wireless technologies, components, products, software, standards, tools, systems engineering, and applications. In short, the price/performance of wireless is now more than competitive with wired solutions in the majority of installations and applications, but dominance here (or even in price or performance alone) doesn’t tell the whole story – it’s also important to evaluate total cost of ownership, and often in comparison to that of wire. Let’s look at the two key components of TCO, CapEx and OpEx, in terms of both wired and wireless outdoor networks. CapEx normally includes any hardware and software involved, but also non-recurring, one-time costs related to planning, installation, initial configuration and functional verification, and required documentation. It may also include any upgrades or additional equipment required over time as the configuration and its mission grows and changes. OpEx includes any ongoing and recurring expenses for network operation, including network management, monitoring, troubleshooting, maintenance, repair, training, and support, as well as any recurring fees from service providers. Let’s examine these in more detail for both wireless and wired networks, exploring the significant differences between wired and wireless installations with respect to these cost components: •

Wired Installations – CapEx here can obviously be so enormous as to be completely cost-prohibitive - running broadband cables outdoors, especially if they are to be buried underground, is ferociously expensive. While the wire itself is not expensive, the planning, legal work, and physical installation are. For this reason, wired broadband service is usually desirable only when appropriate cable already exists, or when a carrier or utility is planning to install broadly-deployed service. Moreover, changes and additions to the installation can similarly be very expensive indeed. Breakeven and payback analysis requires a long timeframe often stretching into decades. In fact, the installation of wire in rural or sparselypopulated areas may have no good ROI scenarios at all. On the other hand, enduser capital equipment requirements are minimal, with recurring (and often significant) service costs the primary expense component.

•

Wireless Installations – Contemporary broadband wireless equipment, on the other hand, is very easy to install. Unlicensed products require at most local regulatory approval and installation by an appropriate professional, and the use of

Wireline vs. Wireless: Exploring TCO in Outdoor Applications

3

Farpoint Group White Paper – June 2009 licensed frequencies involves a straightforward application process usually completed by the equipment dealer or installer. A variety of network topologies is possible, from point-to-point (or P2P) links that duplicate the function of wire but with only a fraction of the installation expense otherwise involved, to point-tomultipoint (or P2MP or “star”) configurations providing extremely cost-effective distribution with again minimal installation effort. Outdoor wireless meshes can also be used to address almost any configuration requirement or network topology, and minimize backhaul expense. All of these configurations duplicate the functionality of wire with a tiny fraction of the installation expense otherwise involved. And while outdoor wireless equipment is very reliable, some installations will specify redundancy in the form of cold standby (spare equipment that can be quickly powered up in the event of a major failure), warm or hot standby, or even fully-redundant configurations often with automatic battery backup power supplies. The rapidly-decreasing costs for wireless (and power) equipment seen in recent years make this option surprisingly affordable. And operating expense can be extremely low, as no fees to service providers are involved, and wireless management and operational-support systems have reached the point where they are as sophisticated as those typical in wired networking. Centralized management consoles can provide a high degree of automation with respect to configuration, re-configuration, troubleshooting, monitoring, and reporting. Little knowledge of the intricacies of radio is required, and many outdoor wireless systems will automatically re-configure in the very unlikely event that an infrastructure node fails. Such failure is, by the way, often the result of a wire being inadvertently unplugged in a wiring closet or a power failure (automatic backup power supplies are often desirable); these are also the primary reasons for failure in a wired network as well. But should greater damage to the wired plant occur, as is sometimes the result of accidents (vehicles colliding with utility poles, for example), construction (the infamous “backhoe outage”), or even errors during wired network reconfiguration, correcting the consequential outage can take some time, with associated lost productivity for wired users. To sum up: with respect to outdoor networks, the key alternatives are the use of wireless capital equipment with minimal associated installation, configuration and operating expense, or wire with a limited capital cost (assuming the use of a service provider) but potentially very high operating expense. And we assume here that wired installation performed without a service provider would be a very, very rare event indeed. Building a TCO Model Our overall approach to the analysis of cost elements for both wired and wireless in terms of an evaluation of TCO is to perform the cost comparison just noted. As we’ll see in the examples below, this model allows us to enter values for the various cost elements involved in each case and determine a breakeven value - in other words, we can quickly see the option of choice. Note, however, that we cannot directly factor in the productivity benefits from wireless in every case. This is a “soft-dollar” element and needs to be Wireline vs. Wireless: Exploring TCO in Outdoor Applications

4

Farpoint Group White Paper – June 2009 evaluated on a case-by-case basis. However, irrespective of the value assigned to productivity, a hard-dollar cost analysis is usually all that is required to show the advantages of a wireless solution. Note also that we don’t attempt to determine an absolute return on investment (ROI) here. Such a calculation is highly industry-dependent, and also a function of the accounting rules used in a given case. We assume, however, that a network must provide some ROI, however measured; otherwise, networks would not be fixtures in essentially every enterprise, commercial, industrial, and governmental setting. And, regardless, minimizing TCO is an excellent strategy for minimizing ROI no matter how it is measured. A Couple of Examples We put our model to work in two real-world deployments that show how outdoor wireless is increasingly the option of choice. Bee Cave, a small city in Central Texas, is a good example of the benefits of cost-effective outdoor broadband wireless. The application here is fairly straightforward - the point-to-point interconnection of municipal buildings with the requirement for significant throughput. Bee Cave opted for Proxim Wireless’ GigaLink® radios, which are rated at 1.25 Gbps, providing both high performance today and significant headroom for future expansion. The installation is currently being used for VoIP services, and expansion to provide backhaul capabilities for Wi-Fi in a nearby park and a popular outdoor music venue are in the works. With a population of only about 2,000 (as of 2007), this installation shows how even very small communities can gain the benefits of broadband via outdoor wireless. The question at hand, however, is how cost-effective such a broadband wireless implementation really is. The analysis in Table 1 shows a savings over a wired solution of almost $18,000 in capital expense alone. Adding in operating expense over five years shows an additional cumulative net savings of $6,500 – a financial benefit that will continue well into the future.

Table 1: Cost model for Bee Cave, Texas. Source: Proxim Wireless Corp.

Wireline vs. Wireless: Exploring TCO in Outdoor Applications

5

Farpoint Group White Paper – June 2009

A second example can be seen in the unlicensed WiMAX installation performed in Alston Moor, a town of about 2,500 on the coast of northwestern England. The Cybermoor network, as it is known, covers essentially all of the town’s population, and offers such services as telemedicine and remote learning in addition to Internet and Web access. The choice of WiMAX in this case was made to enable tight control over quality of service (Wi-Fi is based on best-effort), and the system operates in the 5.8 GHz. unlicensed band to avoid interference from 2.4 GHz. devices. The radios used in this case are Proxim Tsunami™ MP.11 models.

Table 2: Cybermoor analysis. Source: Proxim Wireless Corp.

What’s interesting here is that it’s not always necessary to evaluate operating costs over time in order to build an effective business case based on TCO – the savings in capital expense alone are overwhelming in this example (see Table 2). Running cables to and within this small rural community would have been entirely cost-prohibitive. In addition to a CapEx cost savings of more than US$700,000, the community reports that they are seeing an increase in property values and new-business growth, and an ROI of more than 25 times the amount invested. Clearly, such a return would not have been possible with wired service, even it were feasible to wire the entire town. And the best part: the cost- effectiveness of the equipment chosen has allowed residential end-user prices for true broadband connectivity to be less than nine pounds (about US$14.50) per month for up to 12 Mbps service!

Conclusion and Recommendations As we have seen in this document, it is possible to objectively analyze the cost components of both wired and wireless outdoor broadband solutions for a given networking requirement, and easily determine which alternative is preferable. It’s always best to begin, of course, with a careful statement of project parameters, including such elements as performance, number of users, types of applications, and quality-of-service requirements. Operating expense must be estimated and amortized over the useful life of a given installation unless the capital expense calculation is definitive on its own. It goes without saying that such financial elements as cash flow and opportunity cost (the cost of the project compared to the next-best alternative - which is sometimes doing nothing at all!) must of course be considered, and it is also usually desirable (but often difficult) to factor in soft-dollar cost savings in some form so as to more accurately estimate ROI. Farpoint Group has found that wireless is increasingly the option of choice in outdoor settings. Just as wireless LANs already match the performance of wire with a Wireline vs. Wireless: Exploring TCO in Outdoor Applications

6

Farpoint Group White Paper – June 2009 convenience that is otherwise unmatched in the networking world, outdoor wireless links, especially in areas where wired services are very expensive or otherwise unavailable, are the obvious choice. Outdoor wireless solutions are also very competitive even in locales where wired services are plentiful, again amortizing capital costs over the useful life of a given installation in comparison to monthly recurring service costs. With an (often overwhelming, as we saw in the examples in this document) advantage in price/performance, wireless is today becoming the default or even primary or only access for many in large-scale outdoor deployments, and on a global basis. It is nonetheless important, especially in today’s economic climate, to use every dollar as carefully as possible. We believe, for this reason as well as the others we have discussed in this White Paper, that wireless will continue to build its influence in networking to the point of dominance across the planet, and for essentially every application – indoors and out.

Wireline vs. Wireless: Exploring TCO in Outdoor Applications

7

Ashland MA USA 508-881-6467 www.farpointgroup.com [email protected] The information and analysis contained in this document are based upon publicly-available information sources and are believed to be correct as of the date of publication. Farpoint Group assumes no liability for any inaccuracies that may be present herein. Revisions to this document may be issued, without notice, from time to time.

Copyright 2009 – All rights reserved Permission to reproduce and distribute this document is granted provided this copyright notice is included and that no modifications are made to the original.