Software-defined networking (SDN), network functions virtualization (NFV), and the cloud have the potential to revolutionize communications service provider (CSP) infrastructure and service offering. But operations support systems (OSS) setups often hinder service agility and slow time-to-market, keeping CSPs from reaping the full benefits of their investments in next-generation technologies. A new white paper from Hewlett Packard Enterprise (HPE) says true infrastructure transformation is not possible unless CSPs also revamp the way they operate their networks and create and deliver services.
HPE believes OSS provides the link between business outcomes and NFV innovation. NFV proof of concepts (POCs) are maturing into commercial deployments, but the reality is that most CSP infrastructures will remain a hybrid mix of traditional, physical network infrastructure and virtualized (NFV-enabled) infrastructure for the foreseeable future. The white paper argues it’s becoming increasingly urgent for CSPs to transform OSS to cover both virtualized and traditional infrastructure.
The paper, “Reimaging OSS for Service Agility in Hybrid Networks,” outlines HPE’s new approach to help CSPs manage services across hybrid networks. Traditionally, service fulfillment processes depend on a workflow methodology that executes instructions sequentially to achieve desired outcomes. As services become more personalized and adaptable, the workflows become exponentially more complex. Now, the types of services that need to be created are becoming increasingly personalized based on multiple parameters such as user policies, applications, and network conditions. The workflow methodology is falling short.
HPE’s new “dynamic descriptor” approach gets rid of the classic workflow-driven orchestration where service behavior is hard-coded into the workflows. Instead, it uses a declarative language to describe the service, its relationships, and behavior (policies). The engine then takes the service descriptors, looks at the current state or configuration of the service, and builds a runbook at run-time to achieve the desired state.
This approach bridges physical and virtual environments by providing a common model for data, relationships, and behavior of objects across both the physical and virtualized infrastructure. With a model that can be used for service fulfillment and service assurance functions, service onboarding becomes much faster, and service designers can recompose products while accounting for complex product behavior.
The paper highlights HPE Service Director, a key component of HPE’s NFV solution that uses the dynamic descriptor approach and pairs the domain OSS with the NFV layer. Designed as an end-to-end service management entity, HPE Service Director couples the virtual and the standard production worlds and provides a homogeneous model toward the BSS layer. Initial pilot deployments show a drastic reduction in the time required for foundational processes like VNF onboarding and network service graph creation — from months to hours.
Read the white paper to learn more about how HPE’s new approach to end-to-end service operations can vastly reduce firefighting and drive down the cost of delivery for CSP networks.