The virtualized service world of SDN and NFV will be a catalyst for service innovation, including service mashups, service blending and more, as operators try to provide innovative services anytime, anywhere and across any device.
In the world of software-defined networking (SDN) and network functions virtualization (NFV), balancing customer experience with resource optimization will be crucial as operators focus on providing a consistent, unified user experience across multiple devices, networks and customer touch points. Whether it's virtual network or physical services, quality of service (QoS) will ultimately be the key differentiator.
In order for operators to execute blended service models successfully, a predictive analytics-driven approach to service management, including network resource management and end-to-end service management with an eye towards the customer experience, will be essential.
Enabling innovative services
While most of the conversation around this topic has centered on service orchestration, the pivotal role that predictive analytics will play in the context of end-to-end service management is often overlooked. Next-generation service management platforms need to borrow principles of self-organizing networks (SON) and feature self-healing capabilities to proactively take care of device constraints or provisioning issues based on service quality before the issues impact the customer experience.
The dynamism of a next-generation hybrid virtualized service world demands self-monitoring for anomalous events in the network and the ability to diagnose and fix those issues dynamically. This enables load balancing and optimizes network resources as well, which provides economic benefits to service providers.
Next-generation service management platforms must be driven by predictive analytics to arm service providers with the ability to monitor and track real-time changes to the network on a per-application and per-subscriber level while pre-empting service degradation by constantly monitoring, measuring and maintaining service metrics in real-time. An embedded predictive analytics layer should conduct "what if" analysis and forecast demand, workload allocation and assign probability to future events, and trigger pre-emptive actions.
For example, if a virtual network function (VNF) is exceeding design capacity, at what point should the VNF instance be scaled up? Or during periods of network congestion, when should the VNF move to a different data center and traffic be rerouted? Or when the cost of capacity delivery goes above a predefined threshold, at what point should VNFs be allocated to a lower cost solution?
Operators also want to understand where to instantiate VNFs based on business metrics, such as operating margin, server cost or energy efficiency. In the dynamic world of virtualization, operators will have no choice but to use big data and predictive analytics to collect and correlate the business metrics needed to drive decisions with regard to where to instantiate/move VNFs within the NFV infrastructure.
For enterprise customers, when there is a need for more bandwidth to maintain service level agreement levels or quality of experience KPIs, predictive analytics will work closely with fulfillment and assurance systems to provide that information to fulfillment platforms on a pre-emptive basis. This way, additional virtual machines can be provisioned on the fly or the initial virtual machine can be moved to a higher capacity server or to one that is less weighed down by other applications.
Improving service delivery
In the world of virtualization, service management based on predictive analytics is absolutely critical for streamlined service delivery. Predictive analytics need to be a key ingredient of an operator's service management fabric, which needs to operate in real-time and work closely with service fulfilment and assurance. There are many reasons for these requirements:
- When the service is first provisioned, how does the system understand which resources to use? What are the current loads on the impacted systems and networks?
- How can changes be made to provisioned services in order to maintain a positive customer experience?
- How can performance parameters be measured in real-time and help to provide customer-centric service experiences?
- How can VNFs be measured accurately to make sure that they can scale properly based on defined policies?
- How should one monitor provisioned policies in real-time so that Virtual Machines (VMs) are allowed to move based on configuration policies?
- How should one measure the key parameters relevant in the virtualized world, including measuring latency between network functions, performance of connecting links (VNF to VNF, VNF to legacy, etc.), bandwidth, QoS, latency, SLA measurement, intra-VNF component link measurement, SLA compliance management and verification, measuring threshold alerts based on SLAs, etc.?
As NFV is adopted into mainstream production, virtual machines and VNF instances will start growing exponentially, making the management of such instances increasingly complex. The role of predictive analytics in the context of holistic service management will become mission critical as the need to visualize and measure the performance of a complex and heterogeneous architecture spreads across multiple domains.
— Ari Banerjee, Senior Director, Strategy, NetCracker,
special to The New IP