Service orchestration is important, but there's a lot more to it than being able to allow customers to quickly self-provision connections and bandwidth. Orchestration should also mean being able to rapidly detect and resolve connection problems.
I read a lot about service orchestration, and the top vendors and industry organizations that talk to me about it are manifold. There's the MetroEthernet Forum (MEF), with its Third Network, and Lifecycle Service Orchestration vision. There are companies like CENX, Cyan, and Tail-f (now part of Cisco). All too often, the messages are good, but repetitive: Customers are sick of waiting weeks or months for new connections. They want to be able to do their own moves, adds, and changes. They want to have MPLS service or Carrier Ethernet to have the agility of, say, the ubiquitous Internet.
Obviously, every enterprise customer wants those orchestrated self-service carrier services now, and it's a mumble-mumble gazillion dollar opportunity, says some analyst.
Don't get me wrong: I believe every word those guys are saying. Orchestration is key. The MEF Third Network's LSO is happening. (See my thoughts about LSO here.) Network service orchestration is required to support advances in cloud computing, for leveraging the genuine advances of Software Defined Networks (SDN) and Network Functions Virtualization (NFV), which offer not only the ability to reconfigure the network quickly, but the capability to push applications out to the network edge.
To quote from the Tail-f page on network service orchestration:
Service providers are under extreme pressure. On one hand, surging traffic volumes and sprawling infrastructure have made managing their networks more expensive and complex. On the other hand, increased competition is lowering revenue and forcing service providers to introduce new services more quickly. The result: Profitability is shrinking. The keys to fixing this situation are to simplify operations (which ultimately reduces network costs) and replace the traditional, slow, high-touch provisioning model with a real-time, automated approach.
I agree. Everyone agrees. What happens next, though? Orchestration is good when times are good. Let's talk about when times get bad. In that regard, I was delighted to see news this week from CENX, talking about its customer success with Ericsson. CENX's software is called Cortx Service Orchestrator, and instead of the usual happy talk about how Cortx helps Ericsson provision new services faster than a speeding bullet, CENX focused on rapid problem detection and resolution. To quote:
CENX's Cortx Service Orchestrator delivers an intuitive graphical dashboard to identify performance degradation (e.g. frame loss, latency and jitter) and simplifies the isolation of faults, significantly reducing Mean Time To Repair, increasing network availability and enhancing customer experience.
The capabilities of CENX's Cortx Service Orchestrator allow Ericsson to identify degradation of Ethernet circuits before they begin to significantly impact end users of the network. It provides service delivery teams with the ability to measure Local Exchange Carriers and Alternate Access Vendors' performance against key performance indicators and manage them in line with service level agreements.
That's not something that we read about very often – using orchestration to help carriers be proactive in fixing things. Doesn't every telco prefer to say that its Carrier Ethernet circuits are bulletproof, and every packet happily finds its way home? Kudos to Ericsson for talking about how it solves real-world problems, and to CENX for making it clear that orchestration means much more than simply letting customers set up new end-to-end carrier services in minutes instead of weeks. It means keeping those services working.
This article is published as part of the IDG Contributor Network. Want to Join?