MADISON, Wis. -- Heavy-duty smartphone users often worry about data plans, minutes, and battery drains. Network operators fret over ever-growing data volume, bandwidth, and capital expenditure. Hardware designers sweat processing power, scalability, and the increased signaling their next-gen boxes must handle.
The analytics on certain mobile apps’ misbehaviors -- its unintended consequences on networks, consumers’ data plans, and hardware designs – has been “a bit of a blind spot” for all parties in the mobile world, Josee Loudiadis, director of Network Intelligence, Alcatel-Lucent, tells EE Times.
Alcatel-Lucent released Wednesday, April 30, its Mobile Application Rankings report.
Asked about the industry’s reactions to this report, Loudiadis says, “Although we’ve shared this report only with a few network service providers so far, they’re loving it.” App developers called the report “an eye-opener,” while hardware box designers are calling it “huge,” she adds.
Stressing that Alcatel-Lucent, too, is a hardware box vendor, Loudiadis says, “Our hardware guys use it to understand the future of networks, [their] growth rate, and the increasing load their systems need to handle.”
To create the analytics in this report, Alcatel-Lucent measured two key metrics: “data volume” and “signaling consumption” -- each demanded by different mobile apps. In its view, both metrics are critical in probing the impact of the mobile apps’ misbehavior.
The report said, “Data volume drives the service providers’ bandwidth-related capital expenditures and the consumers’ data usage fees.”
Meanwhile, “signaling,” defined as mobile apps’ frequent interaction with the network, “uses spectral, hardware and processing resources in service providers’ networks,” said the report, thus depleting “battery life in consumers’ mobile devices.”
Further, “the ratio of signaling and data volume exposes each app’s efficiency.” The resulting intelligence offers vendors the chance to optimize software designs and use network resources proficiently.
But how exactly do these analytics help companies in the mobile world improve their business? Loudiadis shared with us one high-profile case: Facebook.
Facebook’s ambition to expand its business to the mobile world is well known. At the Mobile World Congress earlier this year Facebook CEO Mark Zuckerberg urged mobile operators to zero-rate Facebook. He told operators that Facebook can bring the Internet to the next 2 billion users, concurrently boosting operators’ profits.
The argument wouldn’t have flown had Facebook been stuck where it was a year ago. It turns out Facebook last year learned that its app, unexpectedly, increased overall signaling load as much as 10% overnight with the company’s new software release. Such a spike spells bad news for operators, because they know it will cost their networks. Facebook at that time had no idea how this happened, let alone what to fix.
When Alcatel-Lucent blogged about some of the findings of the company’s analytics last year, Facebook saw the blog post, and “they called us,” said Loudiadis.
In their efforts to trace the culprit, the Alcatel-Lucent team members and Facebook staff found the signaling spike was caused by misbehavior by the app on the Android platform. Five months later, Facebook released a new Android version, which resulted in not only reversing the signaling effect, but actually improving it.
The moral of the story? “Having visibility of the apps’ impact on mobile networks is key to reducing the cost of delivery and understanding apps that could be packaged to the benefits of all parties,” the report concluded.
Framework for network impact rankings
The report laid out the following framework -- divided in four quadrants -- illustrating how the team categorized network impact rankings.
(Source: Alcatel-Lucent Wireless Network Guardian (WNG), Network Analytics)
- Apps in the High Impact quadrant (in red) are those that could be optimized for both traffic and signaling consumption.
- Apps in the High Volume quadrant (in blue) are targets for traffic optimization.
- Apps in the High Signaling quadrant (in yellow) would benefit from service provider-app developer collaboration aimed at optimizing signaling usage.
- Apps in the Low Impact quadrant (in green) are perfect for packaging with a data plan using flat- or zero-rating or other innovative options.
Next page: Key findings