YOU ARE AT:5GMEF on 5G deployment considerations and automating lifecycle service orchestration

MEF on 5G deployment considerations and automating lifecycle service orchestration

Because there is no single standard 5G network architecture, there is a lot for an operator to consider

Communications industry association MEF has roots in mobile transport and mobile backhaul. But, according to MEF Board Member Rami Yaron, as new use cases are introduced in the 5G era, carrier needs are evolving, causing MEF (formerly Metro Ethernet Forum) to expand its services.

“5G is much more than just another technology,” Yaron told RCR Wireless News. “There are lot of things that will be introduced with 5G and you have more use cases. There is a huge demand for IoT, and also, people are talking about low latency and a reliable network. This is where [MEF sees the] need to do more than just backhaul, so now MEF works on fronthaul as well.”

Yaron also explained that as users continue to connect more and more devices to a network and the number of use cases for cellular connectivity increase, carriers are now seeking to establish more dynamic, agile networks.

To address this need, MEF is developing Lifecycle Services Orchestration (LSO) specifications with open APIs to automate the entire lifecycle for services orchestrated across multiple provider networks and multiple technology domains within a provider network.

“MEF is making sure that as a service provider, you can have a dynamic, agile network and you can do it with partners,” explained Yaron.

According to the MEF website, LSO enables service providers to transition from a silo-structured BSS/OSS approach towards flexible end-to-end orchestration that unleashes the value of SDN and NFV.

And because there is no single standard 5G network architecture, there is a lot that an operator has to consider when deploying a network.

Yaron said that taking into account how many frequencies a carrier will be using on their 5G network is “extremely important.” Carries have to ask how much spectrum they can use and how much they are able to reuse it.

Carries should also ask themselves what traffic they are running.

“Is it IoT, which requires very low bandwidth? Is it a critical function, meaning network reliability is the most important thing? Or is it consumer traffic?” he said.

Further, the type of 5G applications that an operator is interested in running is often influenced by location. In the U.S., for example, carriers want mobile broadband because of concerns over degradation of customer experience due to congested networks.

“So the U.S really needs 5G to cover more users and to provide more bandwidth,” said Yaron, “but other places might not have this issue, so they might focus on IoT applications for 5G or on developing smart communities, and their network deployment considerations will be different. And based on the application, you can define the architecture.”

Lastly, Yaron said carriers should consider site acquisition when establishing a 5G network. Whether or not acquiring new sites will be a challenge for a carrier, will influence network topography.

Another critical piece of the 5G puzzle is network slicing, which is the optimization of network and spectral resources by providing unique, personalized data pipes that support a particular service or customer. Network slicing will deliver better customer experiences, as well better support for vertical-specific enterprise and industrial customers.

“Different applications have different requirements, and carriers want to slice their network so that each slice can support those different requirements. You can do this with one network and compromise the connectivity of some applications, or employ a multinetwork, which is costly, or you can do network slicing,” Yaron explained.

Yaron also discussed the importance of automating network slicing, saying that this is where LSO comes into play.

“With LSO, an application can pick the right slice that its needs and can have the right delay for the transport that it needs. This is how we envision it with SLO, which can allocate slices for the backhaul or fronthaul, and can also talk to the orchestrator of the mobile because it has to be end-to-end because that’s the main idea,” he said.

ABOUT AUTHOR

Catherine Sbeglia Nin
Catherine Sbeglia Nin
Catherine is the Managing Editor for RCR Wireless News, where she covers topics such as Wi-Fi, network infrastructure, AI and edge computing. She also produced and hosted Arden Media's podcast Well, technically... After studying English and Film & Media Studies at The University of Rochester, she moved to Madison, WI. Having already lived on both coasts, she thought she’d give the middle a try. So far, she likes it very much.