A few years ago, I had lunch with a Bob Olshansky, a visionary at GTE Labs who was regaling me with the benefits of moving all voice traffic over to IP data lines. His vision was of a great convergence of technologies; everything was going to be part of the great internet revolution -- data, fax, voice, or whatever. Unified messaging was the next "killer technology" in the heady days of the Internet bubble. Now, in the post telecom meltdown and dotcom era, what has happened to the VoIP revolution?
"We installed VoIP at our new facility in Reston, VA. It works perfectly, and it cost the same as it would if we had installed a traditional POTS switch. I would definitely recommend it if you are building a new facility because I see it as an investment in the technological future, but the ROI is less certain in a situation with existing infrastructure," says Stan Tyliszcak, Advanced Systems Manager, General Dynamics.
VoIP is most definitely still with us, but how it is used and who uses it have changed. It still has the potential to revolutionize our understanding of telecom and IT infrastructures, but there are some things to consider and watch for. Let's take a look at the pros and cons of VoIP technology for your network and address some valid questions you might be asking about the technology. Questions like how will installing a VoIP system affect the infrastructure of my network and IT systems, and is this a mature technology that I can take advantage of now, or will I have to wait to realize the benefits?
Definition of VoIP
What is VoIP, and can the technology benefit my company? Quite simply, Voice over IP allows telephone calls to be transmitted over IP data networks or the Internet instead of over public switched networks. "In a VoIP network, digitized voice data is highly compressed and carried in packets over IP networks. In a public switched telephone network (PSTN), for example, a dedicated 64 Kilobits per second (Kbps) end-to-end circuit is allocated for each call.
Using the same bandwidth, a VoIP network can carry many times the number of voice calls as a switched circuit network with better voice quality," says an International Engineering Consortium VoIP tutorial. "A VoIP network carries voice traffic cheaper than a switched-circuit telephone network because IP-telephony networks make better use of available bandwidth."
The technology is certainly here, but have we seen the benefits of better voice quality and cheaper services? Many of the industry leaders seem to think that while the potential is still there, it's mostly unrealized at this point.
"On the 'pros' side, it can be cheaper. If you have plenty of data bandwidth [fat pipes that aren't typically congested with data], odds are that VoIP can squeeze in without adding more bandwidth. Thus, it's free bandwidth. That's the obvious win, of course. And it can be a major cost-savings compared to separate voice calling to some foreign countries, notably Latin America. (Bandwidth to the EU is so cheap it doesn't matter much any more.)," says Fred Goldstein, President of ionary Consulting, a company specializing in telecom and VoIP issues.
Tyliszcak echoes Goldstein's sentiments. "The real benefits to VoIP are better integration with unified messaging, call centers where the computer screen can act as the telephone, and other combined voice and data applications. It also offers more robust failover backup for systems survivability." Of course, business continuity and disaster recovery are high on many company priority lists these days.
Chris Ward, Senior Director of marketing at iBasis, a wholesale VoIP service, notes, "VoIP is still a viable business opportunity because the cost of bandwidth has gone down faster than the price of international long distance. Over 10% of the international telephone traffic is now VoIP."
It sounds like VoIP has some real advantages. Well, perhaps, but on the minus side, "The international market is already used to lower quality service and is very price sensitive. VoIP is still new technology, so the quality issues have not completely shaken out," continues Ward. Goldstein also reports that, "On the 'cons' side, there are a bunch of issues that don't get enough press," including:
- Inferior voice quality
- Network congestion issues
- Unreliable equipment
- Organizational politics
- Incompatible revenue models
Goldstein lays out the problems, "Voice quality is inferior. That can impact productivity, because people have to talk slower, or worst case, it's like 'bad cellular,' and you don't want dachshunds or mattress dudes showing up! How inferior is largely a design issue; VoIP with heavy compression saves bandwidth, but at serious audible [quality] cost.
"[VoIP also] doesn't work well on congested data networks. If the data network is fairly busy, it will discard packets, which will impact voice quality. Data retransmits; voice doesn't. This is probably worst on the costliest routes, of course. Another issue is that real TCP data slows down in the presence of congestion, while streams such as VoIP don't. VoIP traffic can crowd out data if it's a large fraction of the traffic on a given link.
"Data gear is less reliable. Telephone networks have a 'five nines' tradition. Even PBXs don't have much downtime. TCP/IP networks have more downtime, almost by design, although now the equipment is getting more reliable."
Tyliszcak adds, "There can also be political problems caused by merging support organizations that traditionally never worked together. Between the telephone, network and systems groups, it is not always clear who is responsible for the equipment and support."
VoIP follows the data business process model rather than the telecom switch model. You bill by the size of the pipe, not by the amount of traffic on that pipe. Software to track and bill for calls by the minute doesn't exist.
The most important disadvantage according to Goldstein, "Non-IP voice is usually cheap enough! The bandwidth cost of domestic calling is almost always below a penny a minute; LD carriers give decent deals to even moderate-sized businesses who don't want to run their own voice networks. VoIP's putative savings are thus solving 1983's problem more than 2003's."
How Can I Take Advantage of VoIP?
So, is it worth the investment? The answer is, as usual, "it depends." VoIP is evolutionary, not revolutionary, technology. If you are moving into a facility where you need to completely build out your infrastructure, then installing VoIP equipment makes sense. It can be very competitive with traditional POTS switched networks. Some companies put a VoIP blade in an existing POTS switch. They feel that it gives them the best of both worlds, but I suspect that they might be hedging their bets a bit too much and not benefiting fully from either technology.
There are some other ways that you can take advantage of VoIP technology. In a "green field" installation, you have a number of attractive choices that can give you more network flexibility. One possibility is eliminating the wires entirely. There is nothing preventing you from installing VoIP over wireless 802.11x networks. "The security algorithms do introduce some latency, but because of the short distances they shouldn't be a problem," says Tyliszcak.
Unified messaging is another example. Unified messaging is one-stop shopping for all your voice mail, email, and fax communications. Unified messaging lets you retrieve all your messages with one phone call or visit to a website. While unified messaging sounds great, the technology has not entirely jelled yet. It can be costly and complicated to implement, and the services have not been perfected, including problems with voice to text translation. Unified messaging is dependent on the VoIP technology so as that application matures, there will be more incentive to implement a VoIP solution.
Recently, Cisco has been investing heavily in VoIP for the enterprise market. They are sponsoring a major trial with the city of Houston, TX and are replacing the entire municipal government telecom service with Cisco VoIP equipment. They claim that it will save 50% over the costs of maintaining the old equipment.
Of course, it isn't clear whether or not replacing Houston's old system with a more modern switched network would result in similar savings. As Chris Ward, notes, "Enterprises' IT departments are real pragmatists. They will not invest in a technology unless there are demonstrated cost savings or substantial improvements in the quality of service."
Here's a checklist of things to watch for when considering whether to invest in VoIP:
- Is your traditional POTS equipment (if already existent) overdue for replacement?
- Are you planning on implementing unified messaging and other integrated voice/data services?
- Do you have enough bandwidth for all your data and voice needs?
- Do you have good connectivity to the Internet?
- Do you have a substantial international presence or communication requirements?
- Does your company have multiple locations?
Beth Cohen is president of Luth Computer Specialists, Inc., a consulting practice specializing in IT infrastructure for smaller companies. She has been in the trenches supporting company IT infrastructure for over 20 years in a number of different fields, including architecture, construction, engineering, software, telecommunications, and research. She is currently consulting, teaching college IT courses, and writing a book about IT for the small enterprise.