Keeping Quality of Service High as LTE Traffic Increases
Today’s “always on” world is all about serving a generation of mobile users with an insatiable appetite for the most current devices. These same users also demand immediate response time, whether they’re reviewing critical business dashboards or simply catching up on Facebook.
With consumer expectations at an all time high, wireless carriers are under constant pressure to keep the network humming and the cost of operations down. At the same time, the challenge of keeping infrastructure costs down continues to grow. According to the latest Cisco Visual Networking Index, global IP traffic has increased more than fivefold in the past five years and is on track to triple by 2019.
Overall, Cisco predicts global IP traffic will increase at a compound annual growth rate of 23% from 2014 to 2019, ballooning from an impressive 1.1 zettabytes annually (or 88.4 exabytes per month) in 2016 to a new milestone of 2.0 zettabytes per year in 2019 (the equivalent of 168 exabytes per month). The global Internet of 2019 will handle roughly 64 times the traffic volume of the Internet of 2005, and the number of devices connected to IP networks will be three times higher in 2019 than the entire global population.
Beyond the sheer volume and scale of the modern landscape, the composition of global Internet traffic is also undergoing change. In 2014, 40% of total IP traffic originated with non-PC devices. However, Cisco is projecting a very different story by 2019. By then, non-PC devices like TVs, tablets, smartphones and machine-to-machine (M2M) modules will dominate, comprising 67% of all IP traffic. In addition, wireless devices and skyrocketing demand for video present additional wrinkles to the traffic flow. The same Cisco report found that by 2019, traffic from wireless and mobile devices will exceed their wired counterparts by a third, and nearly a million minutes of video content will be transmitted across IP networks every second in that same time period.
The breadth of transformation is enormous, and carriers who are unprepared to operate efficiently and effectively in this new environment will be at risk for significant customer churn. More than half of U.S. consumers (51%) switched service providers in 2013, compared to only 5% in 2012, according to Accenture. A more recent survey from Accenture revealed that 46% of U.S. consumers believe they are more likely to switch providers compared to 10 years ago. Along with price, network performance and purchase experience now influence customer satisfaction far more than the type and number of devices carriers offer, according to industry analysts.
So what are carriers doing today to address these rapid changes? Fueled by increasing data demands and customers’ requirement for faster and more responsive performance, they are aggressively investing in an array of technologies and infrastructure enhancements, including mobile backhaul partnerships to bolster speed, reliability and consistency.
In fact, demand for mobile backhaul in the U.S. will increase 9.7-foldbetween 2011 and 2016, according to research firm iGR. And, while mobile backhaul services are a critical piece of the modern network and are central to a carrier’s ability to maintain performance and quality of service, the relationships present some visibility and management challenges. Carriers need to closely monitor their backhaul partners to ensure customer expectations are being met. At the same time, those providing backhaul services need transparency into their own network performance to maintain agreed-upon Service Level Agreement (SLA) benchmarks and avoid costly penalties.
Along with mobile backhaul partnerships, carriers are evolving their networks in other ways to create the bigger, faster and more reliable pipes that can support the higher levels of traffic and more robust data types. A mass migration to LTE (Long Term Evolution), a packet switching, all-IP network standard, is well underway to help carriers meet consumer demand for bigger and faster connections to mobile devices. Yet, while infrastructure is being overhauled to support LTE, there’s also a need to co-exist with multi-technology hybrid networks. This requires carriers to interwork and maintain visibility across a variety of technologies, protocols and network resources.
To balance their LTE infrastructure investments, carriers are looking for opportunities to lower costs, increase average revenue per user (ARPU), and in turn, maximize profits. Boosting network capacity without exponentially increasing the cost curve is a top priority, particularly as they build out their mobile backhaul infrastructure. Those same goals are front and center as carriers work the other side of the backhaul equation – positioning their own bandwidth and network coverage to other wireless providers, which may have coverage gaps in their specific regions.
Amidst all the complexity of this modern and interwoven IP landscape are higher-stakes SLAs, particularly as they relate to mobile backhaul operations. To ensure compliance with the strict guidelines and increasingly standard SLAs, carriers are scrambling to put tools and practices in place so they can validate the performance of their services. Ensuring quality of service (QoS) and quality of experience (QoE) is not only central to carriers’ success, it’s critical for driving adoption of LTE services. Therefore, clear visibility into QoS and QoE metrics becomes essential to helping combat escalating customer churn while optimizing operations and nurturing additional service revenues.
As part of their quest for end-to-end visibility into network performance and service availability across complex architectures, carriers typically measure a variety of key performance indicators (KPIs), depending on the layer and specifics of their SLAs. From the standpoint of Layer 2 and Layer 3, they’re looking for a window into packet loss or frame loss to help determine if the infrastructure is stable. Networks consistently dropping frames indicate potential problems that can lead to service degradation, just as overloaded links and delayed packets are symptomatic of higher latency issues that might result in performance bottlenecks and service delays. Jitter and frame delay variation, determined by measuring whether packets are arriving consistently and in order, is another metric that carriers should be continuously monitoring. This information helps gauge whether data is coming over the network in a choppy fashion—a phenomenon that can have serious implications on performance and ultimately on a carrier’s ability to deliver on its SLAs.
Most carriers are tracking these metrics and have ready access to key data points. However, there are complications when it comes to getting true end-to-end visibility. Most current SLAs are enforced across different network mediums, including SONET, Carrier Ethernet, Microwave, xDSL, DOCSIS and a host of others. Industry standards like Y.1731 and TWAMP are allowing both the mobile carriers and the backhaul carriers, if they’re different, to validate the links and handoffs.
Therefore, to gain true visibility into LTE network performance, carriers need monitoring technologies that leverage these standards to be protocol independent. In this way, they can gain access to a universal KPI and dashboard that spans all transport protocols, allowing them to pivot from performance metrics to related flow records and log data, all from a single screen. In contrast, most traditional monitoring and management environments are built on separate, non-integrated platforms, requiring carriers to toggle between screens and dashboards
Completing the Puzzle
Knowing what to measure is the first step in network performance transparency and a key tactic for ensuring customers are satisfied with the speed and quality of their mobile experience. More than raw data points, however, carriers are best served by having actionable intelligence and insights that can direct them toward making the best adjustments and investments to optimize traffic patterns, ensure network performance, and even adjust service offerings to maximize their revenue streams.
Legacy performance monitoring platforms aren’t cut out to solve these problems. The majority don’t cover the entire multi-vendor and multi-medium infrastructure, which severely limits carriers’ ability to capture insights into potential service issues or effectively troubleshoot problems once they arise. This incomplete data set also impedes their ability to do adequate capacity planning as their LTE networks come onboard, which reduces efficiency of operations.
Without a holistic view, carriers are also unable to understand exactly how individual customers experience activity on key web sites and applications. Perhaps Facebook is running slow in Kansas, but has a normal response time in greater Seattle. Without a window into end user experience, carriers are at greater risk for customer churn, especially in this uber-competitive climate, where users won’t think twice about jumping ship to a company offering a mobile experience that more closely matches their needs.
In contrast, next-generation performance monitoring delivers a consolidated view across mediums and multi-vendor environments, removing visibility gaps. As a result, carriers can see exactly what’s happening in the network, why it’s happening, and to whom it’s happening to, so they can quickly make adjustments. Such a holistic monitoring tool should also accommodate the hyper-scale infrastructures that are synonymous with today’s always-on society,-- allowing carriers to quickly and seamlessly expand the monitoring effort without concern that the larger scope will degrade performance.
In addition, carriers need access to performance data at a highly granular level. This information helps explain complexities in traffic patterns and pinpoint areas that could benefit from investments or changes that might not be obvious at first. For instance, knowing exactly where and when traffic peaks occur could highlight a specific problem or a reason that site A is always twice as busy as site B. This insight could help a carrier determine whether a configuration issue was causing a problem or if the scenario was just expected behavior.
Consider this capability applied to a real-world example: a carrier that upgraded a specific link multiple times to keep up with increased demand in a certain region. When an executive finally questioned why the network team was submitting frequent requests for additional capacity at one location, but not others, they turned to a more capable monitoring platform to troubleshoot the issue. The carrier learned they had a configuration issue that was sending a significant portion of traffic to a single geography. By readjusting the traffic flow, they were able to address the bottleneck cost-effectively and avoid an additional capacity investment.
This type of granular intelligence can be leveraged in a variety of other ways as well. With an exact handle on how fast links are filling up, carriers can be proactive about capacity planning, as opposed to scrambling at the last minute. In doing so, they can actively prepare network infrastructure prior to events that will likely spike bandwidth demand — like the Pope’s visit to the U.S or the FIFA World Cup in Brazil—to ensure consistently and quality of service. The same idea applies to congestion issues that might impact SLAs or service quality glitches like jitters and crackles.
The ability to drill down and turn standard KPIs into actionable intelligence allows carriers to make the necessary adjustments or be confident that performance levels will remain acceptable.
Other factors worth monitoring at such a granular level:
- Backhaul prioritization - to better determine Class of Service (CoS) prioritization for voice, streaming video, email, file transfers, etc. and plan for future CoS growth in specific markets.
- Backhaul congestion - to identify oversubscribed classes and sites, as well as traffic discarding patterns.
- Backhaul path service quality – to ensure backhaul providers are delivering on their promise by measuring jitter and latency by markets or sites.
With a next-generation performance monitoring tool, carriers can also analyze traffic patterns to determine which services are growing or shrinking. For example, what percent of traffic is LTE versus HSPA/EVDO? Once the carrier gets to the point where 90 percent of traffic is LTE, they can decide if it’s time to save costs by eliminating 2G/3G services and antennae from their towers. That actionable intelligence helps them effectively tailor their product portfolios, get new services up and running quickly, and ultimately, acts as a springboard for maximizing revenue.
By opting for a next-generation tool that supports “big data” performance monitoring and management, carriers gain access to a rich data set that fills in visibility gaps and allows for true end-to-end monitoring of network resources. With such an approach, carriers aren’t blindly troubleshooting network performance or retooling resources after a problem occurs. Instead, they’re armed with the intelligence to proactively make network changes or operational adjustments that ensure service levels are consistently met in a cost-effective way.
The bottom line is, stakes have never been higher, and the network infrastructure never more complex. Carriers looking to make their mark in today’s always-on, mobile world need turn routine monitoring into actionable intelligence that takes the customer experience to new heights – all while opening doors to new revenue streams.