Emblasoft’s active monitoring solution ensures QoS and SLAs for multiple 5G differentiated and virtualised services, end-to-end in real-time.
5G is set to support a diverse range of new and complex use cases and offer differentiated services. While the possibilities are exciting, this complexity also brings challenges in terms of assuring performance, Quality of Service (QoS) and reliability.
It means that, as the 5G network evolves, network operators face additional challenges in supporting an increasingly complex number of interconnected network layers and services requiring dynamic provisioning. In addition, networks are embracing cloud technology, virtualisation and network slicing, which add to the complexity of launching, monitoring and assuring services.
Each service has its own set of specific parameters and metrics that need to be measured. For example, the KPIs that need to be monitored for a customer streaming an HD video to their smartphone will differ markedly from an IoT-enabled device, such as a wind turbine. In turn, the KPIs and performance requirements for autonomous cars and smart cities will be different again.
Likewise, the introduction of network slicing means that each slice will have a unique set of metrics and KPIs to monitor and assure. In a recent survey of more than 300 companies conducted by Dimensional Research, participants said that network complexity is growing and increasing the incidence of network outages, while two-thirds state that monitoring solutions fail to predict most issues.
Traditionally, networks were very static entities and, even if they did require strict network KPIs, measuring service performance was not particularly demanding from a monitoring point of view. At the same time, services were completely aligned with network topology, so monitoring the performance of the physical network – rather than the quality of the service itself – was all that was necessary.
Passive monitoring involves capturing and analysing live network traffic and statistics at a specific point in the network. It may simply include the periodic monitoring of byte and packet transmit and receive numbers or, more typically, detailed, non-real-time analysis of signalling protocols, application use and performance. Passive monitoring is ideally suited to in-depth traffic and protocol analysis in a ‘post-event’ scenario, and for gaining in-depth insight into customer QoE.
Conversely, active monitoring emulates realistic user behaviour to determine potential network performance, rather than examining actual users and data. Emulation occurs in real-time at set intervals to collect and analyse simulated performance data.
A significant benefit of active monitoring is that it provides real-time, complete visibility into the network – although it is simulated behaviour it allows network operators to identify potential problems before a new product launch or before any issues affect users.
In essence, active monitoring takes an automated, proactive approach to network troubleshooting by highlighting potential problem areas, while passive monitoring enables real-world, in-depth analysis of network performance on a post-event basis.
The ideal scenario, therefore, is to use both passive and active monitoring, essentially providing the best of both worlds. Furthermore, in today’s dynamic, virtualised 5G environment the need for constant software upgrades and evolution requires a DevOps approach, embracing continuous development (CD) and continuous integration (CI).
However, one of the challenges of supporting active monitoring in a virtualised environment is that network operators need to be able to test and assure network performance themselves.
When networks were largely hardware-based, network equipment manufacturers would test and assure network equipment and performance. However, in a virtualised environment, and with many services running on private [MG(1] networks, network providers are no longer prepared to support network assurance. This places the onus on network operators and private organisations to perform testing and service level assurance for their own networks.
Emblasoft’s Active Monitoring solution consists of software-based agents that can be installed – dynamically, on-demand – in any network domain to generate real traffic sessions and measure QoS for differentiated services and applications.
Furthermore, Emblasoft Evolver can simulate and monitor realistic traffic at both the user and control plane, and fully supports DevOps practices with extensive automation capabilities.
A wide range of use cases can be supported for end-to-end testing (measuring latency and errors in user applications such as RCS and SMS), and for video or IoT applications.
The nature of network testing and monitoring is fundamentally changing, and CSPs need to embrace active monitoring to ensure QoS and to meet SLAs across a spectrum of cloud-hosted services and applications in both public and private networks.
To find out more about how Emblasoft’s Active Monitoring solution can help you assure KPIs, QoS and service performance in a virtualised and dynamic environment – while fully supporting DevOps practices and test automation capabilities – download our latest white paper here.
Emblasoft Test & Measurement AB
120 30 Stockholm, Sweden