In less than 24 hours anyone who is someone in the VoIP business will be headed down to San Jose for the Spring VON show. They will be busy showing off their wares, many of them falling in the “me too” category. And if history is […]

In less than 24 hours anyone who is someone in the VoIP business will be headed down to San Jose for the Spring VON show. They will be busy showing off their wares, many of them falling in the “me too” category. And if history is any indicator, them most of them should and must talk about outages at major VoIP service providers, the current #1 Vonage and Lingo. Last year AT&T’s CallVantage had similar issues. These outages are a wake-up call for the entire VoIP industry, which so far has coasted on good will generated by consumer interest in the fledgling technology. The outages and lack of reliability of the VoIP services could turn consumers off, which in turn could slow down the momentum.

The outages are also a reminder that despite what VoIP loyalists might say about the technology, VoIP service lacks the dependability of PSTN network. I don’t expect VoIP to become as reliable as 100-year-old PSTN, but still I would have expected a cellphone network type reliability. I had been expecting these kind of outages for a while, because according to networking history, every-time you hit one million subscribers, the network complexity increases manifold. In our optimism for VoIP we forgot that back in 1990s, the IP networks would crash all the time. Are we seeing that kind of unreliability we should expect from VoIP networks? How reliable is the network infrastructure provided by the likes of Global Crossing and WilTel?

It is time to step back and take a critical view of the state of VoIP service. A reader reminds me that a similar outages were common place when the PSTN went digital, and back in 1990 even AT&T voice network went down when new SS7 software was installed. Vonage and Lingo could be having similar problems. But there is a bigger issue facing the entire industry: centralization of VoIP resources in few locations. “There was a reason for hierarchical networks in voice and data,” a source tells me. More than Vonage and Lingo’s problems, I think we need to focus on the back-end products. How scalable are soft switches? How do the gateways hold up in face of increased usage? What do you folks think?

Lately I have started to worry about all the hype has overtaken the reality of VoIP. Look at the number of companies at VoN this year – its is easily three times the 2004 total. My VoIP press release count has gone from five a day to 50 a day. And rising. Dot.com of late 1990s has now become dot.VoIP.

You’re subscribed! If you like, you can update your settings

  1. Fazal Majid Sunday, March 6, 2005

    The reliability of conventional voice will start gradually declining for two reasons:

    1. Telcos don’t want to put any more capex than strictly unnecessary in legacy equipment, and protocols like SS7 degrade catastrophically under load. France Telecom has had two major outages in the last 5 years due to SS7, for instance.

    2. Due to bloodletting at firms like Lucent or Nortel, the institutional knowledge about the software in switches has mostly disappeared. The same thing 5 years ago caused AT&T’s Frame Relay network to stop completely because Cascade (acquired by Ascend, acquired by Lucent) had lost all the staff who had worked on the switches in the first place. As new engineers get in on the complex systems, they don’t know why certain coding decisions were made and you end up with complex bugs that are not triggered in the lab but occur in real-world conditions.

  2. Fazal, thanks for the reminder – actually a lesson – on the coming problems in the telecom networks. what i am trying to say in the post up there is that VoIP needs to step up to the plate and offer similar quality if it needs to fill the coming PSTN void. I think that is the critical issue.

  3. steve kamman Monday, March 7, 2005

    Also worth remembering that expectations for quality are likely less stringent as people rely more on wireless. Just a personal observation.

  4. Irwin Lazar’s “Real-Time” Blog Monday, March 7, 2005

    VoIP: Over-hyped?

    This week is VON, perhaps the leading trade show focused on service provider VoIP topics, issues and trends. And not coincidentally, perhaps the most hyped show in the VoIP space. In his Blog, Om Malik’s questions whether or not VoIP

  5. Irwin Lazar Monday, March 7, 2005

    Keep one thing in mind – a phone may be necessary for survival in life and death situations. Will people trust an Internet phone when they need to call for emergency help? What happens the first time the press picks up on a death from someone who tried to call 911 during a Vonage outage and couldn’t?

  6. Abhishek Lahiri Monday, March 7, 2005

    The main attraction for VOIP is that it is cheap. As long as calls made over VOIP are cheaper than calls made over POTS VOIP will have a market. Technological advances and Knowledge gathering will happen over time. Give it a couple of more years and I predict VOIP will replace POTS in most places unless the governments clamp down on VOIP with new regualtions and legislations.

  7. Evidently the feared evnt of 911 failure happened recently. In my opinion this is an artificially created problem because all of us are thinking either/or. With the advent of TV, radio didn’t disappear; with TV and cable news, print media didn’t disappear. For sure lots of restructuring happened in both radio and print. I for one will advocate keeping both the networks and wireless to boot. Then I pick and chooose the appropriate network for the appropriate scenario. The “intelligence” at the end point will help me to select the appropriate one.

  8. The problem isn’t that VoIP is intrinsically unreliable, or that VoIP networks have hit a tipping point in complexity, or that resources are concentrated or localized; it’s deeper and more subtle than that. Network reliability requires a culture of Reliability, Reliability, Reliability, and then some more Reliability, which is 180 degrees opposed from being a fast-moving, low-cost provider of disruptive technology for voice communications. Either approach is OK – but if a carrier markets itself as a “Phone Company”, they’re setting up expectations of Phone Company Reliability. If their culture is that of a fast-moving low-cost provider of disruptive technology, the mismatch will lead to problems.

  9. James Dennis Tuesday, March 8, 2005

    The question I always ask when confronted by someone touting replacement of SS7 communication technology etc…

    “Interesting ‘magic box’ you have there, would you rely on it for emergency calls? ”
    The answer is invariably “no thats not quite what is intended for”. Or once amusingly a “Lord no not a chance” from someone touting the 5 nine’s availability of their solution….

    Same point as illustrated simply.

  10. It’s more than just reliability in terms of up time for VoIP, I am concerned that it becomes an easy target for a terrorist to take out large blocks of our communications at once. Remember that the PSTN has geographical diversity and there are priority emergency lines that get through even in major toll network outages, as well as local calls. I am not aware of any active interest by Homeland Security in this yet, but as VoIP increases it’s penetration I’m sure that will need to be addressed as well.

Comments have been disabled for this post