Mobile service providers have a real challenge on their hands. The rapid adoption and global dominance of the smartphone as the prime tool for communicating and connecting has pushed mobile networks to the edge. With dominant all-IP communications, and demand for instant access to thousands of apps, fast download speeds, and wide usage of delay sensitive services such as VoIP and video, make it extremely challenging to prevent network disruptions and even outages that may affect millions of users.
Network outages are certainly hitting mobile operators where it hurts: Spirent Communications reported that operators are spending $20 billion a year to deal with service continuity issues.
The Obvious Causes of Service Outages
In February 2016 Spirent Communications polled 54 global mobile operators about the main causes of network disruptions. The findings, not surprisingly, pointed to busy hour congestion as contributing factor in outages, and network congestion and overload as the most common cause of network outages and degradation. Network failure was the cause of the most severe outages and degradation. Another very important finding was an increase in the number of issues taking 48 hours or more to fix. Not good for operators with millions of users who are all too happy to switch to a competitor as soon as any service issues arise.
Hidden Causes of Service Outages
The poll lists more general causes of network disruption. But could there be more specific issues or less obvious culprits affecting service availability in mobile networks?
An interesting study at the University of Chicago on the causes of unplanned outages at online service companies classified the cause of 294 (48%) of the 516 cases studied, as “unknown”. The team could simply not determine the cause. In complex mobile networks, finding the root causes of service outages is also likely to be very difficult.
The Unpredictability of Configuration Changes
In today’s complex IP-based communication networks, many factors can affect service integrity.
Take network configuration. Software upgrades and patches almost always have an effect on system configurations in different parts of the network. These changes are very difficult to keep track of, and there may be no telltale signs of the effect these changes will have on the network’s overall stability, service availability, and disaster recovery readiness.
Knowledge driven, Infrastructure-wide Approach
The only way to keep track of all these changes is to deploy analytics driven, infrastructure-wide solutions. There are many monitoring tools that warn network managers of network issues as they crop up. A step further in the right direction would be a solution that warns managers in advance of how configuration changes may affect the network. These proactive detection solutions perform automated analysis, comparing configurations against vendor recommendations, known best practices and specific organization’s policy to pinpoint and alert managers about risky deviations across their IT environment. This way, IT managers can fix issues before they affect network uptime.
By implementing a proactive detection solution, IT managers can also divert more resources to ensuring quality of service and to expanding the infrastructure to meet growing demand.