We’ve been here before with LTE. A major surge of demand for network capacity swept the telecom industry with exploding appetite for immediate access to applications and services. Carriers, finding themselves in a race to capture revenue from new services, responded by introducing new technology as fast as they could as capital investment in access and core networks rose exponentially. As speed of network deployment became a key differentiator, operational efficiency and scalable network management struggled to keep pace. The time has come again with 5G to deliver to a much higher promise of differentiated services and quality.
With vendors rapidly transforming their software packaging to micro-services and disaggregated user/control planes, Communications Service Providers (CSPs) are investing further in edge-distribution across distributed clouds. Network Functions Virtualization (NFV) shared resources, multi-vendor architecture, multi-tenant slicing, elastic workloads, cross-cloud migrations, traditional and container execution units, are just a few operational considerations ahead. As the race beings again, the challenge, lies in planning and executing faster than ever before, and the operations management approach needs to be a well curated design from the onset.
VMware vCloud NFV takes a full lifecycle approach to operations management by incorporating core attributes such as simplification, multitenancy, proactivity, and efficiency across various aspects of a large-scale telco operation including Service onboarding, Service Launch, Performance monitoring, Issue isolation with machine learning, Dynamic and predictive optimization, and Capacity planning and business management.
Here, we take a closer look at this approach and discuss how integrated operations intelligence in VMware vCloud NFV facilitates simplicity, 360-degree visibility, and automation, in an agile telco network deployment.
As a global operations management function, the operational intelligence platform provides a single line-of-sight across the virtual infrastructure, overlay, and underlay networking and is fully expandable with service-awareness and physical underlays.
VMware vRealize Suite – Operations, Network Insight, and Log Insight, together make it easy to operate and manage the next generation 5G distributed NFV clouds.
Take a 5-stepped journey through the operations lifecycle management:
Onboarding Service Operations – While the 5G ready cloud infrastructure will isolate compute and networking for each customer context, the CSP operations manager will graphically define and discover the service and resource boundaries for their customers, add custom SLA triggers on top of the many existing alerts, create custom dashboards, discover and plan security micro-segmentation profiles to protect east-west and north-south traffic, and add closed-loop integrations using the workflow construction interfaces to notify a service orchestrator.
Service Launch – Once the services are up and running, monitor their health and ensure SLA’s are within check. Watch for critical alerts, capacity constraints, performance degradations, anomalies and more in the cross-tier landscape of services, virtual machines, data centers, clusters, and hosts.
Dynamic Optimization – Want to ensure services can be offered in an agile cadence, with the same stringent reliability and quality requirements? Leverage consumption patterns and available resources across hybrid clouds to optimize workloads continuously. The just-in-time operational intelligence is fed back into the NFVI Distributed Resource Scheduler (DRS) as well as north bound triggers to a service orchestrator to achieve predictive-class avoidance.
Issue Isolation – So you received an alert, what next? Follow the issue isolation orchestrated drill-downs into the services’ compute, storage and overlay and underlay networking. Use the automatic machine learnt intelligence to find those needles in the haystack in a single log analysis interface. See something you want to put under future observation, create an alert and dashboard with the click of a button.
Demand Planning and Expansion – Over-provisioning was the way to solve for capacity. Dynamic demand monitoring is feed into an intelligent capacity planning engine which can not only keep capacities in check but also allow you to reclaim unused resources and create what-if scenarios for future demand forecast and capacity exhaust. Evaluate the cost of private cloud operations and consider economies of a public cloud.
vCloud NFV Operational Intelligence @ MWC 2018