Given the increased complexity of networks and the overall service delivery system, meeting high performance SLAs requires the unification of operations across different domains and operational environments. Meeting high performance SLAs mandates cohesive visibility and fault and performance analysis across the ecosystem. The ecosystem consists of the virtual layer, the physical infrastructure layer, the application layer, and the end customer services. It includes a complex network that’s massive in scale. To successfully deliver high performance and meet customer expectations, you must distill this complexity down with data streams and analytics that can easily spot problem areas, analyze for root cause and not just symptoms, and rapidly identify the populations impacted by the problem.

Today, there is a siloed approach to service assurance. This slows down responding to, detecting, and fixing any customer and service-impacting issues. Consider this example reported by a Telco:

In a pre-paid environment, customers were trying to send SMSs, consume mobile data, make phone calls. This requires authentication and communication with the provisioning side to make certain credit is available on the account. There were authentication failures, purse checks, and faults occurring at the VNF layer. Two different organizations were troubleshooting the problem which eventually ended up in a virtual war for resolution. Failed VNFs were causing a problem – it was reporting that many customers didn’t have enough credit on their account to consume the services they wanted. Bad for the customer. Bad for the operator as well.

With the introduction of AIOps, technology organizations have a framework for bringing the fault and performance side together, across service operational units, managing the triage in a more automated way, and then directing the incident to the right automated fix technique or to the right fix team.

Operational Efficiency

Operational efficiency demands more automation which will also propel the industry to the next level – self-healing and autonomous networks. If we think about this journey as six steps, the industry sits between step 1 and 2. To reach the desired state of automation, visibility across the ecosystem is foundational. As long as organizations view the network from silos, observability across all layers will elude organizations, making it more difficult to achieve an automated state.

Choosing the right combination technology, organizations will be able to integrate fault and performance management. Observability across all layers will provide the intelligence to move closer to the goal of autonomous networks. For example, selected AIOps applications provide a unified view of everything that’s happening from all the resources, all the services, all the vendors, all the layers in on aggregated view. This enables the correlation, deduplication, and presentation of performance status on a dashboard in a consolidated incident management inbox.

The automation mandate for service providers will ensure they can discover issues before the customer. This means they reduce meantime to identify incidents and the meantime to know the root cause of those incidents. Ultimately, reducing the meantime to resolve incidents and improve customer satisfaction.

Another requirement to achieve greater automation is to understand probable cause vs the symptom of a problem. To achieve this step, organizations need a way to onboard data quickly, enrich data in real time and correlate across diverse services and topologies of the ecosystem.

Change Management Assurance

These changes support root cause identification as well as identification of the potentially impacted populations. By automatically detecting when network changes have occurred and determining the impact of these changes on the network organizations provide change management assurance. Determining the impact of changes on the network improves the customer experience.

With the visibility, advanced analytics and ontological approach, VIA AIOps also supports the change management assurance processes by automatically detecting when network changes have occurred and determining the impact these changes are having on the network and customer experience.

Vitria and Cisco

Vitria VIA AIOps provides the ability to understand the probable cause vs the symptom of the problem. VIA onboards data quickly, enriches data in real time, and correlates data across various ecosystem services and topologies. VIA uses an ontological approach to obtain the context needed for more intelligent correlation that supports root cause identification as well as identification of the potentially impacted populations.

Cisco Crosswork Network Controller does software defined networking, automated provisioning of services and provides a level of service quality and service problem management for Cisco Domain Networks.

Cisco’s recent acquisition of Sedona Systems provides a hierarchical controller that integrates with Crosswork Network Controller, as well as network controllers from other vendors delivering a unified network view. VIA AIOps is the aggregation layer at the top of the stack, just below the IT service management platform. VIA AIOps aggregates service health information from the Crosswork Network Controller, from third party SDN controllers, and from Sedona, so that you then have a unified view of everything that’s happening from all the resources, all the services, all the vendors, all the layers in one aggregated view that’s correlated, deduplicated, and presented in a performance status dashboard in a consolidated incident management inbox.