Service assurance software vendor making strides in the cloud-native direction while retaining its other options, including VMs.

November 6, 2017

5 Min Read
SevOne Adds Containers to Its Quiver

When it comes to doing service assurance in the virtualized telecom world, SevOne is choosing to cover all its bases. In addition to its alliance with Red Hat to combine service assurance with that company's NFV infrastructure, SevOne also is packaging its components so they can be delivered in containers for cloud-native deployments as well as traditional virtual machines.

In an interview at last week's "OSS in the Era of NFV & SDN" event in London, Tom Griffin, vice president of Strategic Alliances for SevOne Inc. , told Light Reading his cloud provider customers want a containerized approach today and his telecom service providers are moving quickly in that direction as well. The company's SevOne Data Platform, announced in September, will be the underlying structure for a suite of infrastructure management tools, the first of which is SevOne Data Insight. (See SevOne, Red Hat Team on NFV Service Assurance.)

"SevOne Data Insight is our first visible artifact of bringing our solution to being cloud-native," Griffin comments. "We have taken the user interface workflow components of the product and packaged them to be delivered in a container, not moving away from physical application but giving another option in terms of deployment."

Figure 1:

Most telecom operators have initially moved to virtual machines rather than containers, as they adopt network functions virtualization, but "the level of orchestration and automation that they are trying to drive is certainly looking toward the future of cloud-type delivery instead of physical components," he adds.

The challenge will be managing that transition. As Griffin notes, today's NFV deployments are dominated by OpenStack. Its Red Hat Inc. (NYSE: RHT) partner has about 50 telco deployments today.

"That said, everybody is looking toward containers, even though I don't think anyone is ready yet to deploy and ultimately, over time, as with all of these things, I think it will be a transition," Griffin comments.

The attraction of being able to package microservices rather than VMs is compelling from both an efficiency and an operational cost perspective, he adds, or at least it seems to be, ahead of actual proof of the cost benefits. Virtualization's first wave basically packaged the existing physical boxes with their applications into a single VM -- what Griffin calls "a pretty big lump of coal to move around."

"With containers, the efficiency gained is that rather than virtualizing the entire operation and the entire application I can start to package it in microservices," Griffin says. With Data Insight, for instance, SevOne didn't take the entire application but rather its UI components, and put those into a smaller package that can be more easily deployed.

"Fundamentally you have the ability to architect [apps and services] by moving much smaller lumps of coal around, and they require a smaller footprint in terms of resources," he says.

Of course, Google (Nasdaq: GOOG) has been doing this with Kubernetes for some time but for a more limited set of applications that doesn't include legacy services and transport networks, Griffin adds.

"In the telco context, there is inherently a lot more complexity," he says. "If you are a telco, you can't ignore your legacy. That's the big challenge the telcos have. That is why today with the NFV wave we have a number of customers in production but a lot of them are still in pilot or lab type deployments right now."

Want to know more about NFV and Open Source strategies? Check out our dedicated NFV content channel here on Light Reading.

Meanwhile, the growing experience with OpenStack is changing the way that NFVi approach is handled, Griffin notes. Business drivers are pushing telecom operators forward faster, to reclaim revenues being lost to over-the-top business applications, and they are having to take a more practical approach. The pace at which versions of OpenStack are produced is much faster than most telecom players can handle, and that has led to Red Hat's option for an OpenStack release to be delivered every 18 months (versus every six) with a three-year support guarantee.

"That's a pace at which most telecom operators can consume," Griffin says. "Personally, I think that is a good compromise between something dropping every day in open source to change every five years in traditional telco world. We recognize in the telco market if we deploy a solution, we have to support that integration for three to five years, because they just don't move that quickly."

The SevOne exec sees both 5G and SD-WAN as business cases pushing the telecom market forward faster. On the latter front, SevOne has partnered with Viptela, before they were acquired by Cisco, and with Versa. Griffin says partnerships with other VNF providers and a broader ecosystem of players will continue to make sense for his company. (See Verizon Turns to SevOne for SD-WAN Visibility.)

He sees 5G "as the thing that will drive virtualization massively across the network. 5G and NFV are becoming inherently interlinked a lot of what they are doing is with a view to supporting 5G in future."

The broader virtualization space remains a major market opportunity for SevOne, in part because the company is willing to be a critical component in a broader ecosystem and play well among the bigger trees -- like Red Hat or big systems integrators -- as well as VNF vendors.

"We ultimately see ourselves performing a pretty important component but it is only a component," Griffin says. "We are providing insight into the quality and the capacity of the services. The output of what we do is a required trigger for provisioning and policy -- we don't do provisioning and policy but we are a key source of data for all of that."

— Carol Wilson, Editor-at-Large, Light Reading

Subscribe and receive the latest news from the industry.
Join 62,000+ members. Yes it's completely free.

You May Also Like