Orange's CTO spells out need for network API collaboration

Laurent Leboucher, CTO of Orange, spoke at DTTW 2022 about his ambitions to transform the way the communications service provider (CSP) operates networks and delivers services, as well as some of the challenges that network transformation poses.

Leboucher built his career at Orange in IT transformation and played an instrumental role in driving the development of TM Forum Open APIs. He is now bringing his expertise and experience to network transformation and believes TM Forum’s Open APIs and Open Digital Architecture (ODA) have a role to play in enabling network transformation.

Orange is virtualizing many areas of its network, undertaking containerization and moving to microservices. Leboucher stressed that the changes go beyond virtualization and like CSPs such as Telenor or Jio, is embracing a layered platform model.

“It's no longer vertical silos, but different layers … with different economic driving forces,” he said. “It is about disaggregation. We are working with some vendors in different segments of the network.”

A platform, approach, however, will demand the use of APIs between platform layers, believes Leboucher. For example, “we rely on NaaS [network-as-a-service] APIs that will be implemented by the connectivity layer.”

“We are used to exposing IT APIs and we are strongly involved in Open APIs,” said Leboucher. “We want to expose network APIs … - for instance, quality of service on demand, also location. So, different kinds of APIs, which we believe will bring value and that we can introduce on the market.”

However, developing network APIs will require industry-wide collaboration. “We need to agree with other operators; we cannot do that alone.”

“We have to cooperate, and we have to compete at the same time,” said Leboucher. “And of course, this has to be done through APIs.” TM Forum, for example, is collaborating with MEF on applying TM Forum Open APIs to the ordering of broadband services.

The changes afoot at Orange are designed in part to help the CSP offer its customers different kinds of connectivity services that are as easy to consume as a service on demand.

OSS transformation

A different approach to providing network services, however, demands new tools and processes. Orange has put in place an experimental 5G standalone (SA) network, called Pikeo, on which to test new ways of operating networks and delivering services, as well as collaborating with new partners.

Leboucher discussed a number of areas where change is unfolding.

“We need to transform our OSS transform deeply. Especially, we need to have a dynamic network inventory. This is really a foundation,” explained Leboucher.

Orchestration will play an important role in facilitating new ways of working and delivering services.

“In our OSS, we need to introduce orchestration [and] … we want to do it in a way which is … as much as possible homogenous across the different countries.”

A benefit of homogeneity is that it will allow Orange’s network monitoring teams, for example, to interface with field operations in several countries. In addition, according to Leboucher, “we need a workforce management tool … [and] an inventory which is very consistent.”

Continuous improvement

The softwarization of networks also offers CSPs the opportunity to speed up service development and delivery by introducing agile practices already in use in IT.

Although there are differences between the way networks and IT systems are developed and operated, Leboucher explained there are also common practices. He believes networking can benefit from IT culture and practices such as continuous improvement/continuous development [CI/CD].

“We’re putting a lot of effort into CI and looking at CD.” Pointing to the example of Rakuten, which delivers network software features several times a day, he said: “We are moving to a world where we need to deliver software features [for the network] in a continuous way. And it needs to be done with our partners together using the same pipeline. “

Automation plays an important role in helping telcos achieve their network transformation goals and Orange is no exception.

“We have some very, very compelling use cases on automation One of them is related to the quality of voice, how we can … drill down extremely quickly to find the root cause … Thanks to AI, we are able to do it in a matter of just a few minutes; we used to do it in five or six hours.”