Operators will benefit from the flexibility that comes with the ability to apply IT virtualization technology to CPE and network functions. However, virtualization introduces additional complexity, creating the need for an orchestration layer with more sophisticated assurance capabilities, including data center and network analysis capabilities, encompassing physical and virtual resources in real time.
This paper focuses on a valuable use-case, which leverages operations analytics (OA) and machine intelligence (MI) to drive a variety of resource allocations within virtualized networks. This resource allocation applies both within the “cloud” and in the access network.
Within the cloud, OA and MI can be utilized to predict when additional capacity is needed for services such as additional compute to maintain quality of experience (QoE) for a cloud-based guide or additional storage for a cloud-based DVR system. These technologies can also be utilized to predict unexpectedly popular content and pre-position it optimally within content distribution networks. Within the access network, OA and MI can be utilized for congestion prediction and service optimization. For example, trends in PHY performance parameters can be leveraged to predict the need to tweak OFDM/OFDMA profiles for maximal efficiency.
This paper also explores an emerging concept: “Software-Defined Operations (SDO),” which applies Software-Defined Networking’s (SDN’s) separation of data and control planes to key pieces of modern care and operations equipment. For example, if one views technical support calls as the data then the Interactive voice response (IVR) system is the equivalent of a router (and, thus, part of the data plane) and programming that IVR system is, thus, part of the control plane. With that concept established, the paper looks at the power SDO can have when combined with Operations Analytics.
Virtualization creates benefits and challenges, both of which are driven by the new dynamicity of network services, topology, inventory and hardware resources. The combined application of OA technologies and MI can help operators to assure virtualization by providing an automated, real-time approach that learns from data and autonomously adapts to new information, intuiting connections and relationships to proactively detect anomalies and prescribe solutions.