Share this post:
I departed on a trip recently for work after not traveling for a few months. It was an odd rhythm in the summer for me compared to the past number of years. I realized as I traveled across the ocean I normally overlook a number of things as I journey. The same may be true also for the journey of network operations especially when we are buried in the details of it day in and day out. As I have been looking at the ins and outs of operating changes required by NFV/SDN, I find we need dynamic operations for cloud-based networking.
Dynamic Operations enables service providers to seize the value of cloud-based networking. Much is made of the components of NFV/SDN, the hardened environment needed but operating a hybrid network of virtual and legacy networks, a hybrid world of network and IT, and a hybrid world of on-prem and off-prem requires a different type of operational environment, one that is agile, one that can bring together the end to end view, and one that can help cross the diversity of technology and delivery being utilized.
It requires dynamic operations which enables first the ability to simplify, focus and get ahead of the network in a new way. Analytics in all its forms creates this ability to enable this operational model. Analytics brings together the silos of insight across an operational environment of network and IT, on-prem and off, physical and virtual, reduces the information to the essential needing focus by helping to correlate with deep insight about what is going on so a service provider can quickly react to problems in the network, correct and restore service. Analytics also goes beyond correlation and more insight by also enabling the ability to predict what will happen in the network to keep services active before a failure occurs, proactively enabling operations. With the increased complexity of virtual and legacy networks, mixed generations of technologies, and the hybrid world of IT and Network on premise and off premise, the ability to determine patterns and changes occurring at machine speed requires the use of the machine to help think, find the patterns of network behavior and anomalies, and adapt to the changes and seasons of the network. It requires cognitive analytics or the use of the machine to learn these patterns to predict with. Cognitive computing brings these analytic capabilities together with the ability to understand structured and unstructured (or “dark”) data of the network together. It also enables a new level of native language interaction vs knowing discrete commands as it reasons and learns. Together this type of portfolio of analytics enables cognitive operations for a dynamic operating environment.
More than dealing with events
However, dynamic operations is more than dealing with events. In the world of cloud-based networking with NFV/SDN the nature of availability shifts. In a legacy network, it was critical to maintain functional availability to maintain the service end to end. In a cloud-based network, the power of the cloud enables dynamic scalability and the need to focus on component availability shifts to availability of the service. What used to be a standard requirement of five 9’s at the component level begins to fade with the power of the cloud and the shift is to five 9’s of the service delivery. Operations shifts from network operations to a service operations and the move is needed to change Network operation centers to service operation centers or from NOC’s to SOC’s. This change is enabled by the power of the former area of analytics to shift to this type of focus service quality and all of its facets along with the dynamic environment of the cloud.
Moving to operations-as-a-service (OaaS)
The last part of dynamic operations to enable the cloud-based network is utilizing the power of the cloud to deliver the operations for the cloud-based network.
Much is made of moving the network components to the cloud, but are we moving the operations to the cloud? Are we using the power of the cloud (on-prem or off) to deliver the operations needed for the cloud-based network as we combine legacy network and NFV/SDN environments?
Many are hesitant moving to operations-as-a-service and though moving to the cloud can enable this it is much more. Moving operations to cloud native whether on-prem or off allows us to take advantage of the cloud for scaling operations rapidly and as needed. But cloud enables a whole new type of rapid innovation with being able to use micro-services to create iterative changes vs large operation transformation. Cloud also creates an environment that allows for the use of API’s of functions to create new insight, using cloud for the operations innovation needed, and the new insights for focusing on services. Using cloud services reduces cost of delivery but also creates a new level of agility for delivery and agility for incremental changes.
Dynamic operations bring Analytics, a focus on service and the power or agility of the cloud together to enable an operational environment that can provide the service needed for cloud-based networking. Dynamic operations can work at the network, service, and the customer levels to create the value needed for both operators and consumers of the network.
How dynamic are your operations today and for the world coming tomorrow?
Or, are you overlooking this critical area in your journey to cloud-based networking as I have been overlooking much as I travel for work? Leave your comment below.
Visit ibm.biz/Bdrpxi to learn more about the architecture for cloud-based networking.