IT Brief NZ - It's time carriers embrace opportunities from the Cloud

Warning: This story was published more than a year ago.
cloud_enterprise.jpg

It's time carriers embrace opportunities from the Cloud

As enterprises embrace the cloud, carriers have an incredible opportunity to assist with migration, layer on orchestration, and differentiate their services using SDN and NFV.

Enterprises are voracious consumers of bandwidth – and the use of that bandwidth is changing. The bread-and-butter for enterprise connectivity has been WANs, wide area networks that connect disparate business locations.

Carriers, of course, offer a variety of technologies for enterprise WANs, such as the commonly used SONET/SDH and Carrier Ethernet, running over fibre in metro areas, across countries and through undersea fibre links. Those are adequate for yesterday’s WANs, but aren’t enough for tomorrow’s new reality: links to the cloud.

As enterprises embrace the cloud, bandwidth needs change, and so do customer expectations. With cloud applications, branch offices don’t connect to a small number of enterprise data centres; instead, they link to multiple cloud hosts.

Connections must be created and changed quickly, in a matter of hours or days – and the traditional snail’s pace for provisioning new services is simply unacceptable. What’s more, enterprises expect bandwidth to be scalable, safe and secure, and project the corporate network boundary across all of its connectivity.

The requirements of enterprises migrating to the cloud represent a challenge, and an opportunity, for carriers, cloud service providers, equipment makers and the industry. Telcos and others are already embracing new standards, and layering on advanced orchestration and service differentiation, to seize these new opportunities.

The core enablers: IP, carrier ethernet, SDN and NFV

At the heart of advanced carrier connectivity is IP, which is replacing other private-line protocols. “The service providers are tasked with migrating customers,” explains Erin Dunne, Director of Research Services for analyst firm Vertical Systems Group.

“You have many large service providers out there that have said, the old stuff is going away, and we're moving everything to IP. That's a huge task.”

Moving to IP in the carrier realm means offering customers Carrier Ethernet 2.0 (CE 2.0), a wide-area Ethernet service that provides multiple classes of service, manageability and interconnects.

Often implemented on top of MPLS, CE 2.0 carries virtual LANs to provide point-to-point links. Some CE 2.0 services are provisioned to be point-to-point links to provide access to applications; others are configured as multipoint-to-multipoint to handle data broadcasting, video conferencing, and support of multiple data centers.

The other two core protocols for the modern network are Software Defined Networks (SDN) and Network Functions Virtualization (NFV). SDN separates the router’s control plane from the data plane, distributing intelligence throughout the network.

This not only allows the network to be quickly reconfigured without a truck roll, but also allows it to adapt and grow without minimal changes to expensive routers and switches. NFV allows carrier and enterprises to implement additional functions directly on the network itself, such as firewalls and load balancers, without additional hardware

According to a broad consensus amount industry researchers, about a third of service providers are currently using SDN and NFV in their Ethernet or IP VPN service networks or they're planning to do in 2015. About half are planning or considering SDN and NFV deployments further out in 2016 or later. Only 15% of service providers have no plans to deploy SDN and NFV in networks.

Orchestration builds on the care

Enterprises require more bandwidth, flexible connectivity and rapid provisioning, in many cases to support cloud migrations. Protocols and technologies like IP, CE 2.0, SDN and NFV allow carriers to offer those services to support enterprise WANs.

What’s needed to make it work? Lifecycle service orchestration (LSO), which lets carriers quickly provision new services, and also gives enterprises the ability to, in some cases, self-manage their connectivity.

The organisation driving much of today’s standards work behind LSO is the MEF, which also developed Carrier Ethernet. According to Kevin Vachon, COO of the MEF, there are several drivers enabling orchestration:

“Firstly, SDN inherently drives the need to have more complexity and sophistication at the software layer and less at the hardware layer. So as compared to traditional networks where you had a lot of smarts in the hardware, some smarts in the software, SDN is the opposite,” he said.

“You see this sophistication requirement as compared to traditional proprietary networks flipping around, much more sophistication in software. And SDN is happening.”

The second factor driving the need for a substantial service orchestration capability, Vachon continued, is market demand for on-demand and agile services.

“Businesses want to have bandwidth on demand. They want to get network as a service. They want to be able to buy a network capability. They want to be able to get it quickly, they don't want to wait three months.”

Vachon added, “Customers want to have self service portals. If you're going into a portal to change the characteristics of your service, you can't have an army of people in a back room making those network changes on the fly on a live network. There has to be automation.”

“It's no surprise to me in that the industry is moving in the cloud direction,” Vachon said. “Cloud services, cloud applications cannot realise their full potential if the network fabric underneath is not dynamic and agile and automated and so on and so forth.”

Carriers see the value of orchestration, said Amit Sinha Roy, Vice President of global telco Tata Communications, and are beginning to realise LSO will help make networking services more efficient to provide.

“It’s across the lifecycle of the relationship that a provider has with a customer,” added Roy. “Starting from provisioning the service itself, going through to the control, the performance, automation, usage, monitoring, even managing security, pushing policies. It's across the entire lifecycle. Right from the start of the service, the relationship through managing the service and then even switching it off. That's the entire lifecycle that LSO would help manage.”

Orchestration and the cloud

Gint Atkinson, Vice President, Network Strategy and Architecture for network provider KVH Colt, described a real-world scenario about the carrier challenges — and orchestration opportunities — for connecting enterprises directly to cloud service providers. His examples used Amazon Web Services’s Direct Connect service linked directly to an enterprise data centre with a CE 2.0-based point-to-point link, called an E-Line.

“Minimally, on the E-Line going from AWS Direct Connect to the connectivity into the data centre to somewhere else, all the way to the user's data centre, their private cloud and then all the way back into a branch office, there are many Ethernet segments inside there, each one separately provisioned and then orchestrated,” he explained.

“Layer 2 connectivity into AWS can be problematic,” Atkinson continued. “Enterprises need to set up Layer 3 services. Configuring the virtual router is something a service provider can come in, and it’s easiest with end-to-end lifecycle orchestration.”

Taking the cloud connection scenario to the next step requires NFV, he said, when the customer realizes that they're having security issues or a huge slice of that network capacity and CPU capacity is consumed by dirty data. In this situation, an NFV-based security solution from Wedge Networks would solve the problem.

“Now someone in that service chain wants to spin up a security service and Wedge Networks can come in with their NFV and this gets spun up all in orchestration.

It's not just a security function, I'm going to have dozens of other functions that I want to spin up over time. But I really can't do it now quickly and easily because I don't have lifecycle orchestration.”

Atkinson advised, “SDN is not the bleeding edge any more, it's leading edge. And the same thing with NFV, it's not bleeding edge any more. And to really maximise the value out of these capabilities and make new service mixes and a new experience for the users of cloud connectivity, we need lifecycle orchestration.”

The manufacturer perspective

As an equipment maker, not a carrier, Dell has a different perspective – but also sees the value of SDN and NFV, said Jeff Baher, Senior Director for NFV Solutions, for Dell Networking.

“From Dell's perspective, we see NFV and SDN playing out similar ways in the way the laptop in the server industries have,” he said. “Where there is more of a horizontal orientation towards ecosystems, there is semiconductors, there's systems, there's operating systems, there’s applications.”

Baher continued, "On the enterprise side, there's borrowing from a lot of NFV and SDN. We see a lot of large enterprises trying to advance their private wide-area networks with many NFV related technologies. Everyone is coming at it from different angles, leveraging a lot of the same technologies."

Paul McCluskey, Director of Business Development at orchestration solutions provider CENX agreed, but sees that the move to SDN, NFV and orchestration won’t happen overnight due to the huge quantity of legacy networking infrastructure.

“Look at the amount of money spent on infrastructure every year, especially in the U.S., Look at the wireless carriers, tens of billions of dollars. That stuff's not going to go away.  So we know that NFV is necessary,” he said.

“But not overnight. You can't just switch everything off and replace it. All of that infrastructure's been paid for, it's earning money.  So there's a need to not just orchestrate across the new infrastructure but also across the existing infrastructure.”

McCluskey put the right words to the question facing the entire industry: “How do you offer services ubiquitously across your existing and all of the new infrastructure at the same time?  That’s the big challenge for the service providers as we go through this transition.”

Article by Alan Zeichick, Camden Associates president and principal analyst​.

Interested in this topic?
We can put you in touch with an expert.

Follow Us

Featured

next-story-thumb Scroll down to read: