Integrators : Consolidation and Flexibility

Most organizations have put forth mandates to the network and infrastructure teams to consolidate the number of OEM vendors and tools that are currently being used by the operations and engineering teams. But should consolidation reduce flexibility, including the flexibility of choice? We have recently begun to notice that many vendors are trying to take over the data center. We think organizations should be empowered with choice—the choice of what is right for their customers—and not get locked down by anything.

Achieving this flexibility requires training teams in multiple technologies, which adds to IT expenditures. It also requires managing all these devices: upgrading, licensing, and all the expensive “operations stuff.” So, it does make sense to consolidate to a few vendors. But what do organizations do when they identify a new vendor providing excellent functionality that will lower costs after a few years? The cost and time involved in taking on a new vendor is high, and it is not only about purchasing and deploying the new vendor device. It also involves the cost of training users and customers, updating management platforms, and so on.

It doesn’t stop there. What about the other systems? How to get the new devices and their management platforms to work with existing tools? Ticketing systems, workflow engines, patrolling systems, IP address management systems, authentication systems…a nightmare indeed! Organizations having been doing this for ages now.

Integrator

This definitely requires a solution, and it is not something new. It has been a buzzword for a while: yes, it is SDN. With varying definitions of SDN all around, you no doubt believe in data plan/control plane/management plane separation. The controller is intelligent and has northbound programmability. This solves the complexity in L2 and L3. L4–L7 services have always been programmable. Companies like F5 have done well here. Not only have their devices offered northbound programmability, but the BIG-IQ platform is simplifying that now. But the third plane, the management plane, is basically the window into all these controllers and the information they provide. The applications on the management plane should be able to talk to multiple vendors (controllers or devices) and integrate with the organization’s current change control systems and incident management systems. It should also provide analytics, because someone needs to make sense of all that data. The management application needs to be an “integrator.”

This “integrator” management application should provide the right blend of monitoring, management, and troubleshooting. It should have the right hooks into change systems (automation) and should be extremely easy to use. Consolidation and cost saving needs to happen in the management plane to allow the data and control planes to adopt new technologies as they come.

Tags

  • Network Automation
  • Network Infrastructure Automation
  • Network Infrastructure Management
  • OEM
  • orchestration

About the Author

Ashok Kumar B

Sr. Automation Engineer

More From the Author →

Related Articles

Top 7 Benefits Of An Intelligent Web Application Firewall

| 6 Min Read

Canary Deployment Strategy: Benefits, Constraints And How It Can Be Used For Application Traffic Management

| 5 Min Read

How To Mitigate Log4j Vulnerabilities For F5 Devices Using ADC+

| 3 Min Read