A Three Minute Guide to Network Automation Bliss

Friday December 09, 2016

The cloud computing paradigm has been making steady progress in 2016. With the DevOps model making its way from cloud to networking, the business upside of fully automated service architectures is finally beginning to materialize. The associated service agility is expected to unleash new business models that transform the ways in which applications and connectivity can be consumed.

To facilitate these new revenue opportunities, service providers are now focusing on techniques such as service abstraction and network service chaining. The idea is that once the networks have been abstracted into virtualized templates, it becomes easy to deploy and to connect networks in real time as required by different services. While these methods are certainly foundational for network automation, there is at least one more challenge that is yet to dawn on most industry participants.

A high percentage of business-to-business activities in service provisioning are associated with operating Wide Area Networks (WAN). Essentially, WANs are the private enterprise networks that function as the backbone of the digital enterprise connecting data centers, machine networks and various company sites via trusted private networks. Safe consumption of online resources within the private network requires seamless connectivity between the next-generation services and the existing WANs.

While network service chaining and service abstraction allow new networks to be deployed immediately, they do not provide a mechanism for connecting these resources with the existing enterprise networks. A key problem here is that accurate network information is difficult to come by, as larger organizations tend to scatter network information across multiple silos. Typically, an internal network registry team would manage the networks used by the service provider itself, whereas the WAN-related information would be managed by several IP architects in various account teams.

The practical implication of this siloed structure is that service abstraction and network service chaining is not enough when new services are being launched. As the network information is scattered across the organization, it can easily take weeks or even months before the correct information is found and the new services can be connected. This destroys most automation benefits such as scalability and time-to-market, making profitable on-demand enterprise services a distant dream.

To overcome this problem, here's a three-step guide for putting one's house in order:

  1. Organize all existing network information into a single authoritative management system, and establish an automated mechanism for retrieving information from the software-defined world.
  2. Develop a security mechanism that provides the appropriate Authorization, Authentication and Accounting (AAA) for the set of confidential network data.
  3. Implement a unified Application Programming Interface (API) that allows networks and network-related data to be accessed by third-party systems and orchestrators.

As network automation gains momentum, an increasing percentage of organizations have made network abstraction and network service chaining a priority. Yet the lack of organization in basic network management will keep the on-demand enterprise services in a bind. The power of network automation can be unleashed by deploying an authoritative network management solution.



By Juha Holkkola, Co-Founder and Chief Technologist at FusionLayer Inc.Juha Holkkola is the Co-Founder and Chief Technologist at FusionLayer Inc. An inventor with several patents in the US and Europe, he is an advocate of technology concepts with tangible operational impact. Juha is an active proponent of emerging technology trends such as cloud computing, hybrid IT and network functions virtualization, and a regular speaker at various industry events.