Every private and public network operator today needs the ability to visualize not just how the network is performing but also how that performance impacts the business, machines, and people connected to it. This view is known as Quality of Experience (QoE).
Network equipment and network implementations are not traditionally designed to provide end-to-end visibility. The more complex the end-to-end network, the more important it is to focus on QoE. In this article, we examine the needs and requirements of delivering a homogenous service experience over heterogenous network infrastructure.
Public and Private Networks: Dependent and Distinct
Private networks and clouds coexist and interconnect with the large-scale public networks that have underpinned telecommunication for the past century. Yet, each of these networks is designed and optimized for different services and to serve particular customers, subscribers, and users.
Private networks are built to meet needs that public networks cannot or do not currently provide for and generally need to deliver higher-grade connectivity, security and reliability than public networks. For this reason, the quality, reliability, and performance expected of private networks is often of higher grade than that of public networks.
Many private networks are more complex than public networks. Larger private operators—oil and gas, mining, and utility companies, for example—are multi-regional and international. They must integrate and manage networks across several regions and countries. Frequent merger and acquisition activity in these industries also increases the complexity and heterogeneity of networks with multiple vendor implementations.
Machine-to-machine (M2M) networking further increases the levels of complexity in and on the network and drives major innovation in private networks.
Public operators are also responding to these new opportunities by building out IoT networks to provide end-to-end, IoT-grade services and applications. Fleet management, smart cities, smart agriculture, and e-health are examples.
Private and public operators need a way to deliver homogenous service experience
over heterogenous infrastructure and QoS (metrics) tools.
In this context, one common challenge puts pressure on both public and private network operators: effectively managing the performance of end-to-end services and applications despite the complexity behind the connection point. Operators need a way to deliver homogenous service experience over heterogenous infrastructure and QoS (metrics) tools.
Changing the Game of Network Visualization for QoE
Network equipment and network implementations are not traditionally designed to provide end-to-end visibility. Rather, QoS tools and network element management systems (NEMS) monitor individual pieces of equipment and segments of a network. In the NOC, this means multiple systems and screens to monitor; it may also require dedicated individuals focusing on specific pieces of the network.
To deliver scalable QoE and meet the growing demand on IoT networks, the scope and capabilities of network visualization must be redefined.
The more complex the end-to-end network, the more important it is to focus on QoE. QoE visualization can best be monitored and managed on a single pane of glass. One centralized interface must illuminate the experience of machines and people on the network. It must provide homogenous visualization uninhibited by the network’s heterogenous underpinnings. This QoE view must also be available in real-time to the entire organization, including operations, IT, and business decision-makers.
The primary goal or measurement of QoE network visualization is to detect
anomalies before they impact the machines and users relying on the network.
The Role of ML and AI in Network Analytics and Automation
Providing end-to-end QoE in increasingly diverse network environments will require automation. Artificial intelligence (AI) and machine learning (ML) are needed to achieve end-to-end visualization with impact correlation of diverse networks and applications.
For public operators, scale is another concern. With IDC anticipating more than 40 billion connected IoT devices worldwide by 2027, it’s easy to imagine a large telco needing to support 1 billion such devices in short order. Scale at this magnitude will not be possible with current QoS tools, NOC staff, and process alone. Automation assisted by ML and AI is the only feasible means to maintain real-time visibility into an end-to-end network’s QoE with scale and at machine-speed.
For private operators, the criticality of network health and reliability has health, safety, and environmental (HSE) implications that public networks do not. Real-time HSE and business continuity notifications, as well as automated remediation actions, are needed to support and enable ongoing automation throughout these operations and business models.
Automation is also critical to provide the modeling capabilities all operators will depend on to anticipate future needs and how best to address them. This includes knowing where to invest both geographically and in terms of technology (LTE, 5G, SDN, MEC, edge compute, network slicing, etc.), as well as ensuring investments enable improved QoE.
Understanding the goals and current challenges of both public and private network operators, Cheetah Networks PulseView™ is an AI- and ML-enabled QoE analytics solution built on five pillars:
For more information, please contact us or request a demonstration of the PulseView™ Solution.