AT&T ICEs Vendors of Virtual Network Functions.
AT&T revealed two new aspects of its virtualization strategy, both related to virtual network function (VNF) requirements.
The first is a process the service provider is calling ICE. Vendors must use it to introduce and certify their virtual network functions for AT&T’s network.
“The vendor side has improved,” said Chris Rice, SVP of AT&T Labs’ Domain 2.0 Architecture and Design, at the Linux Foundation’s Open Source Leadership Summit. He said the process for vendors to introduce a VNF in AT&T’s network is now down to four days. And that process will take only one day by the end of this year.
Rice said AT&T plans to open source its ICE processing software.
Secondly, AT&T has defined some requirements for VNF onboarding. Originally, the company referred to these requirements as “guidelines.” But Rice said, “Unfortunately, you can get a lot of snowflakes — VNFs that are totally unique — when what you really want is Lego blocks.” Also, the VNF onboarding requirements steer vendors to “not just take functions, virtualize them, and run them on an x86 server, but to build virtual functions so that they’re cloud-native,” he said.
Reaching its Virtualization Goal
AT&T is marching forward with its goal to have 75 percent of its network virtualized by 2020. Rice said there are four key factors involved in making this happen. Two are the VNF requirements and ICE. The other two are AT&T’s OpenStack cloud and its Enhanced Control, Orchestration, and Management Platform (ECOMP). AT&T just recently opened-sourced ECOMP via the Linux Foundation.
The company is doing all of this to handle the exponential growth of traffic on its networks. From 2007 to 2017, it saw 250,000 percent growth in data traffic.
“The way we’ve built our network previously was rack-and-stack,” said Rice. “It’s linear. We needed something that would allow us to keep up with growth. It’s the reason we decided to go down this software path.”
AT&T’s 2016 goal was to virtualize 30 percent of its network, and it achieved 34 percent. Its goal for 2017 is 55 percent of its network.
Rice talked a little bit about some lessons AT&T has learned in its network virtualization efforts.
“When we first started down this path, the same people who built the platform were also building services,” said Rice. But the company soon realized that wasn’t the best idea. However, he added, “The most important set of folks to get involved early on are your operations people and the people building your services so you can get their feedback and get the self-serve capability.”
He said AT&T has an internal goal this year, which it isn’t specifying publicly, that a certain percentage of its services will be delivered in a self-service way.
Have you ever had an OSS or BSS project that’s impacted by poor data quality? The problem of data integrity gets raised on almost every
“One of my favorite things about designing software is designing the defaults. The defaults define the experience for everyone out of the box. And, therefore,
Have you ever been tasked with mapping out the processes for the telco you work for? Either as a way of documenting current state, or
In the last week alone, I’ve had two potent reminders that the world of OSS is simply (pun intended) too complex. This manifests in many
Here at Passionate About OSS, we are exactly that – Passionate About OSS (Operational Support Systems). We’re also passionate about BSS, NMS, or any other names you call the tools that help to operationalise your network.
