Monitoring, managing and monetising QoS
febrero 13, 2020 - Sara Philpott
Service assurance used to be about making sure customers got the right network experience. With 5G, service assurance will take a leap forward as measuring, monitoring and managing quality of service may be tied into service level agreements for the delivery of services.
Like all the latest generations of mobile technology, 5G is primarily marketed on speed. If a consumer lives in a 5G coverage area they’ll be expecting super-fast 5G speeds all the time. For enterprises, if they roll out 5G enabled IoT as well as data services they won’t just be expecting 5G speeds – they’ll be demanding them. For the enterprise market operators need to examine selling service level agreements (SLAs) to be able to guarantee network performance. This is key to rolling out new services on different network slices with different performance requirements, expectations and demand.
All of a sudden service assurance just got interesting as it can be monetized
5G: QoS, QoE and Service Assurance
Service assurance is becoming a key requirement for operators as the focus sharpens on QoS (Quality of Service) and QoE (Quality of Experience) with growing network complexity and service performance sensitivities. The unprecedented speeds that 5G can deliver will drive higher bandwidth use cases, each placing greater pressure on network resources. As an example, with 5G enhanced mobile broadband, a user can download a 15GB full-length high-definition movie in 6 seconds- the same movie takes 4 minutes to download with 4G/LTE networks.
If we consider the use cases associated with autonomous cars, a vehicle travelling at up to 60 miles/hour will be capable of receiving a stop signal when it detects danger with zero (1 millisecond) latency. The car would have moved just a few centimetres before stopping, whereas in 4G, the same car would have moved at least a meter. Insurance companies are obviously evaluating these capabilities with great interest. Furthermore, an explosion of IOT and enterprise applications is expected with the density of connectivity reaching 1 million devices per square kilometre. Operators are investing billions to deploy 5G technology and faced with concerns such as ‘how do we ensure that users are receiving the service they’re paying for, in a highly dynamic and flexible world?’ Or ‘how do we ensure that enterprises hosted on our network are correctly billed with complex 3rd party value chains? The need for effective audit and monitoring systems to protect this investment has never been greater.
Ensuring QoE and QoS requires the ability to monitor end to end services as it traverses the network replicating key measurements from the perspective of the end consumer. Virtual probes and key component function data provide a holistic view of the services and enable the continuous analysis with expected SLAs and thresholds. AI models lend themselves to perform anomaly detection against these service profiles for multi-dynamic transactions and feed automated and closed loop actions to remediate degraded service performance.
Monitoring QoS – Can Uncover Insight to Drive Sales
Due to the sheer number of systems involved in delivering the multitude of complex services offered by operators, the ability to effectively monitor quality of service can be severely hampered. In most instances, an operator is only capable of detecting decline in the quality of a service delivered to customers after those customers are impacted to the point of notifying the operator through customer care agents. Through being able to monitor and predict the quality of experience being delivered to customers, operators are able to identify and resolve issues that customers may be impacted by before they become consistent pain points. Prior to AI’s arrival, network providers typically used some sort of packet filtering, such as deep-packet inspection (DPI), to dissect individual network packets and gather detailed information that could help them find and fix the network problem. However in order to understand the entire flow, metrics must be collected across multiple points through the delivery path and together with feature and data classification on those metrics, a holistic representation of the data flow is generated. This end to end view of the data flows, with the help of AI, can detect problems with 80% accuracy.
On its home networks, Verizon runs automated testing on a sample of 60,000 in-home routers every two hours, to ensure that customers are receiving the speed of service they are paying for. Verizon found that this analysis helped to drive business decisions. Testing showed that the home routers were consistently able to operate at higher speeds than was previously thought. This meant the business was able to market its service as a 1-gigabit connection, where previously it was advertised as 750 megabits. This led to a huge upsurge in sales.
Service assurance and monitoring QoS and QoE are already important to deliver the optimal customer experience. 5G will take this to a new level, due to new bandwidth intensive use cases, the opening up of new uses cases in the enterprise market and the ability to monetise SLAs.
This is an extract from the Openet guidebook: 28 Data Management Use Cases.
BLOG
