When significant real-world ROI results come in from the field, we’re excited to share them. Some recent proof points from our service provider customers are especially compelling. Why? Because they’re about one of your biggest challenges―reducing high operational costs.
Big Reductions in OpEx Confirmed
Before the first human flew through the stratosphere, a space ship was just a fictional concept. So it is with emerging technologies before they hit an inflection point and take off.
Case in point: the automation and orchestration of network services. For the past few years, Cisco has been promoting the benefits of these features as part of programmable networks. The benefits we’ve touted may once have sounded remote and other-worldly, like fiction. Faster time-to-market. Simpler operations that don’t require expensive sys admin time. And especially much lower operational expenditures (OpEx).
Now the results are in and they’re really impressive. The promises have become reality.
Cisco crunched the data on operational processes from one of our major service provider customers that has automated their environment. Specifically, they implemented the Cisco Network Services Orchestrator. Individual service lifecycle processes were analyzed―adding a new customer, resolving an incident, handling a change request, disconnecting a service.
The savings in time and motions came to 60-70%. Overall OpEx savings came to 50-70%. Over five years, that translated to a ROI of 383% and savings of $3 million to $16.7 million from year one through year five for this Tier 3-5 provider. For a Tier 1 operator, the estimated savings over five years could reach $66.7 million!
Significant results indeed.
Where the Greatest Savings Occurs
Aside from looking at the big picture, our study also isolated the cost savings from specific portions of workflows. The green boxes in the chart highlight where network services automation and orchestration had the greatest impacts.
Source: Cisco Business Technology Architecture Group based on averages across mobile, cable, telco and data center service providers.
In the past, sys admins configuring half a dozen sites have had to do so manually. They had to check and recheck configurations. Yet a huge percentage of problems are due to incorrect configuration or moving back to an original configuration.
By contrast, automation and orchestration rely on the data model-driven programmatic configuration of all elements participating in a service—both physical and virtualized functions, traditional and software-defined networking (SDN) management interfaces, operation support systems (OSSs) and business support systems (BSSs)—in a single transaction. That configuration extends from service-level parameters through the automated configuration of every network device involved in the service.
Automated configuration validation, a major area of savings in our study, extends across service chains, all the way from the CPE to the virtual network functions (VNFs) in the data center. The orchestrator knows if it can move on to the ongoing assurance phase of the service lifecycle (or whether it needs to reconfigure or redeploy something that has failed) based on actual measurements of performance at the endpoint.
The providers participating in this study have automated the entire lifecycle of their services, including fulfillment, activation testing, and ongoing monitoring and assurance. They design, deliver, and assure services faster. They have a single, network-wide interface to all network devices and services, both physical and virtual, using a common modeling language and data store. Services are defined using the standardized YANG modeling language and mapped to device models. And everything is automated.
Goodbye to custom coding or service disruptions. Hello to operations in real-time without manual intervention for incredible ROI and OpEx savings.
Automation and orchestration features are part of the Cisco Open Network Architecture for service providers. The architecture represents an infrastructure transformation that leaves past approaches to software, hardware, services, architectures, and operations far behind. We’ll be talking about it and a lot more at SDN World Congress on October 10-14 in The Hague―check out my colleague Frederic Trate’s blog for details.