Telecom operators rely heavily on real-time network intelligence to boost performance, improve customer experience and retention, identify new services, and rapidly capitalize on revenue opportunities. As essential as harnessing the power of this intelligence is to operators, it is becoming more challenging as the volume, variety, and velocity of data dramatically increases. By 2020, global mobile data traffic for enterprise networks and telecom operators will grow 800%. Over-the-top competitors, such as Skype, WhatsApp, and Google, have already adapted to quickly utilize this valuable data – taking over $23 billion in SMS and $170 billion in voice revenue from traditional telecom operators.
To become more efficient, agile, and profitable in the face of these industry trends, 68 percent of telecom leaders are actively implementing or planning to implement Network Functions Virtualization (NFV) in their core networks. Unfortunately, extracting value from NFV presents a challenge as traditional measurement and monitoring tools – such as physical probes – are not optimized for these new, virtual networks. Passive virtual probes, another option, represent a necessary advance over physical probes; however, I believe (along with many in the industry) that they will not be sufficient in unlocking the full value of data in virtualized networks. Although less static than physical probes, passive virtual probes still present scaling limitations, performance impacts, and data collection challenges that, taken together, limit the potential of today’s new virtual network functions (VNFs).
Telecom operators need full visibility of the events and conditions in their virtualized networks along with the ability to perform real-time analytics on network data in order to enable critical business decisions. Let me illustrate this with a couple of examples where network intelligence combined with real-time analytics results in incremental revenue and improved customer experience. First example: continuously streaming data regarding mobile subscriber location is matched with profile data that indicates whether the subscriber has opted into a particular retail promotion. A real-time alert is then automatically triggered when a customer is in proximity of a retail outlet prompting a coupon to be sent to his or her phone, providing an incentive for the subscriber to shop at the outlet. Second example: real-time monitoring and analytics of data usage is leveraged to send proactive notifications to mobile devices whenever subscribers are about to exceed their data usage thresholds, offering easy ways to top up their data. This increases customer satisfaction while lowering call center costs at the same time.
The challenge of executing use cases like the ones described above becomes apparent when you consider the intermittency of data velocity (think of the surge of data that an event like the Super Bowl places on a city’s mobile networks) and the sheer volume and variety of data introduced into networks by new technologies like 5G and IoT. Virtual network functionality needs to be able to dynamically grow and shrink in response to these variable network conditions.
Enter integrated virtual probes. An integrated virtual probe is part of a VNF and, as a result, is instantiated and terminated whenever a VNF is. This tight coupling between probe and VNF allows the number of probes to scale up and down in direct proportion to the number of VNFs, bringing linear scalability into the environment. In addition, integrated virtual probes provide performance advantages by eliminating the redundancy found in passive virtual probe deployments. The ability to grow and shrink probes in proportion with VNFs creates the right amount of probing “surface area” in virtual networks. This enables telecom operators to stream network data in real-time and generate the insights needed to achieve significant cost and time savings. It also facilitates revenue generation from the existing subscriber base, better identification and remediation of network problems, improvement of network quality and capacity, and more.
Dell EMC has partnered with Affirmed Networks on an industry first implementation of integrated virtual probes with Affirmed’s Virtual Evolved Packet Core (EPC) that streams data in an open format to Dell EMC’s Telecom Data Analytics platform. View our new infographic for more information and to see real world examples of what virtual probes and data analytics can do for telecom.
This post was initially published on LinkedIn.