America’s Telecommunications Network Outages: Fewer But More Severe
Today’s Wall Street Journal has an article about the downtime trends in America’s telecommunications network. (The Journal is running a week of free on-line access this week, so you can see it for a couple more days for free; then it will only be viewable with a subscription).
The authors, Jesse Drucker and Amy Schatz, highlight results of a report compiled by the Alliance for Telecommunications Industry Solutions, which looked at the performance of our communication networks in the wake of hurricanes Katrina and Rita. The data shows that the number of outages – or instances that a significant portion of phone service was unavailable to users – has been steadily decreasing over the past 4 years. However, the average duration of each outage instance has been increasing.
The report only covers the traditional “wired” or POTS (Plain Old Telephone Service) network, and not wireless or packet / VOIP networks. Starting this year, the wireless operators are required to report all outages that meet a certain minimum criteria to the FCC, with explanations of cause. However, the FCC will no longer publish this data for security reasons.
What does this mean for telecom vendors? In short, I think the data shows that we (telecom vendors) have some work to do to make the newer technology that we’re putting into networks today more robust. Even in the traditional wired voice networks, most of the telecom products today are based on new packet-based switches and routers. These new platforms have very different operational characteristics than the traditional telephone switches of the previous 20 years: akin to the difference between a blade server and a mainframe computer.
Operators’ networks are also functioning differently and have different types of traffic and use patterns than they used to. Optical transport, for example, allows operators to get a lot more bang (data or voice transfer between locations) for their buck (CAPEX and OPEX). However, concentrating a larger portion of your network traffic on one medium can mean that the impact of a failure is much higher. Of course there is redundancy built in, but the overall impact when even a redundant system fails is higher.
These are just a couple of factors that I believe contribute to the increased outage downtime in the U.S. voice networks today. The telecom vendors must continue to strive not just to deliver the best new features and services to operators, but to deliver the robust, fault-tolerant, easy-to-operate platforms on which these applications reside.
Good article Matt. I had read the brief quickly today, but missed the “meat”. Thanks for pointing it out. Do you think this is also due to consolidation/mergers, new smaller players, increased competiveness, and also just maybe better reporting in a more audit conscious post 9/11, post enron world?
Brad - I think you are definitely on to something with your points. Increased network complexity, newer operators, increased competitiveness which drives down spending on operations all are factors in the matter of outages.
Some of these things won’t change, so we, the vendors, need to adapt our products to work easily in this environment.