As communications service providers (CSPs) continue to transition their networks to a virtualized environment, for the next several years, they’ll be operating their legacy physical network in tandem. The resulting hybrid network environment demands higher efficiency and interoperability in the provisioning and deployment of services as compared to the legacy physical network. Service assurance systems will play a key role in ensuring the hybrid environment will deliver the speed and agility required to provision and deploy new services. The service assurance system is a central system that can align with the speed of service provisioning, be synchronized with the changes in network topology, and enrich the data acquired from multiple sources.
By streaming data in real-time and breaking down operational silos, a central service assurance system can increase agility and interoperability in the hybrid Network Functions Virtualization (NFV) ecosystem. The creation of an efficient, interoperable ecosystem, however, will require—at a minimum—open APIs, dynamic inventory, telemetry sourcing of data, and policy-driven closed loop.
Interoperability
As virtualized and legacy networks are expected to run side-by-side in a hybrid fashion for years to come, service providers are faced with the challenges of multiple API standards, limited or non-existent end-to-end, real-time visibility into their networks, and siloed or poorly connected service assurance capabilities. CSPs’ expectations from a service assurance system include receiving and processing data across the entire network regardless of the surrounding factors—the stage of the network’s hybridity, varying technologies, and vendors, or services and applications supported. In other words, it should consolidate all operations support systems (OSS) functions across the entire hybrid network.
RESTful open APIs can be the first step in solving this NFV MANO nightmare, helping to reduce operational complexity by integrating the various OSS layers as well as the various elements of the CSP ecosystem. The typical ecosystem comprises the physical network, the virtual network, the NFV orchestrator, the software-defined network (SDN) controller, OSS such as provisioning and activation, billing support systems (BSS) such as billing and customer relationship management (CRM), the customer portals (including reporting and dashboarding), the Big Data lake/storage repository and Big Data analytics.
Creating a service partner ecosystem using open APIs can also help in offering and monetizing services to other CSPs and digital service partners in the future.
Here are a few examples of how open APIs can unify the silos of a CSP’s ecosystem, facilitate interoperability and create an environment of collaboration:
- Align the topology and inventory with the service/information models
- Trigger NFV orchestrators and controllers to stream data
- On-demand discovery and enrichment of Big Data for analytics
Efficiency
To build high efficiency in the hybrid NFV ecosystem, real-time operations and data processing are critical. For this, the service assurance system should acquire data rapidly and then perform data analysis for actionable outcomes. These outcomes act as triggers for orchestrators and controllers (NFV resource orchestrator, NFV service orchestrator, SDN controller, etc.), which, in turn, carry out the recommended changes based on established policies for effective closed-loop problem resolution.
The service assurance platform, thus, increases the efficiency of the orchestrators and controllers through:
- Faster rates of data ingestion
- Smarter, contextualized processing of information
- Real-time remediation
Service assurance systems that acquire data faster can efficiently generate precise performance KPIs. Data acquisition based on older methods such as SNMP need to give way to more real-time and secure data-flow methods, such as telemetry, mediation data buses and data flow management.
- Other than sub-second data granularity, telemetry offers measurements such as CPU utilization, memory utilization, I/O operations rate, and more. This ensures efficient and resilient operations.
- Mediation data buses, such as Kafka, allow service assurance systems to acquire data quickly, and to efficiently direct it to the relevant NFV elements (i.e., VNF managers, MANO, hypervisors, etc.), as well as other ecosystem components (orchestrators, Big Data lakes, fault management systems, etc.).
- Data flow management systems such as Apache NiFi can help in smoothing the transformation of data from multiple sources and transferring it at higher granularities.
A service assurance system that is real-time and dynamic in operation will not only bring efficiency to and accelerate the CSP’s virtualization programs during the NFV onboarding and NFV operation stages, it will also serve as a pivot for interoperability between the CSP’s various ecosystem elements and help address the key challenges associated with real-time, dynamic service assurance in a hybrid environment. As CSPs stride forward with virtualization, and adopt digital business models, transforming the service assurance system for greater interoperability and efficiency is a fundamental requirement to maximizing the value of their hybrid networks.
Sandeep Raina is Director of Product Marketing at InfoVista for Service Assurance solutions. Sandeep has worked as a senior executive in the telecom industry for over 25 years, having worked ten years with global mobile operators for network engineering and quality, and over 15 years with OSS vendors to develop and evangelize Service Assurance solutions for new technology networks and services.
