Remember 99.999% Reliability?
After spending literally billions of dollars acquiring spectrum and building out infrastructure, Verizon Wireless’ LTE network has been plagued with outages. On Wednesday the network went down…for the fourth time in just the past three months.
"Verizon Wireless engineers have been working to resolve an issue with 4G LTE service that is affecting some customers' 4G devices," (VZW statement 12/7/11)
“VZW is investigating customer issues in connecting to the 4G LTE data network," (VZW statement 2/23/12)
For the sake of fairness, it should be noted that other major network operators, Telenor and NTT DoCoMo among them, have also experienced well documented 4G network outages.
“On Friday 10 June, Telenor experienced a major mobile network outage, causing all mobile voice, SMS and mobile broadband traffic to be significantly reduced over a period of several hours. Telenor decided to compensate subscribers by giving away all traffic for free during the period 10–13 June,” (from Telenor Q2 2011 Interim report).
These aren’t the type of press releases in which you want your media relations staff to become proficient.
Perception, reality, expectations
So what is to become of these outages? For one, VZW may consider de-emphasizing the “best network” marketing angle, at least for the near term. Avoiding any argument about reliability details or comparison of metrics with the competition; perception is reality. If your customers perceive you have a problem, rest assured you have a very real problem. It matters little if actual reliability is higher than the competition or other services such as cable. To avoid disappointed users Verizon, or any service provider, needs to meet or exceed service levels their customers expect to receive. The “always connected” users of today are “always aware” when a service they have become increasingly dependent upon is not available. As these service outages become more transparent to the general public, the potential for brand damage increases dramatically.
More tangibly, these well documented outages provide validation for some of the strategies and solutions available from some leading network equipment and software tool vendors.
The coming need for a new core control element in the all-IP (from access to core) networks was presented to me in discussions with a signaling industry leader nearly two years ago. They forecasted the coming 4G/Diameter protocol based networks would present challenges similar to those faced by service providers during early SS7 deployments. The signaling requests from the increasingly intelligent and location-aware mobile devices dwarf those experienced in SS7/TDM network architectures. Without intelligent control, this signaling load can quickly overwhelm and bring down a network. Service providers have indicated at least a few of these recent outages resulted from signaling network issues. The new control element (called a Diameter Signaling Controller by Ulticom, Acme Packet and others or Diameter Signaling Router by Tekelec) is designed to provide the load balancing, centralized routing, and policy management needed to provide intelligent central control in a 4G networks.
From discussions with developers and software tools vendors, it is evident there is a resurgence in telecommunications infrastructure spending. This investment trend is not likely to abate soon as mobile subscribers and mobile data consumption continue to rise sharply. VDC believes the increased investment in equipment is subsequently driving a reacceleration of R&D spending and thus creating pull-through business opportunities for embedded software development solution vendors.
More specifically, the growing awareness and impact demonstrated by recent network outages reinforces the potential cost of software failure (operational, future sales, and brand image). As such, we anticipate that the growing importance of this software’s continued functionality will stimulate increased (and renewed) investment in automated testing technologies such as static analysis that offer a means to identify potential defects in the code base.
