Control and User Plane Separation (CUPS) isn’t a new concept in the wireless world but it’s quickly becoming an integral part of the 5G network lingo. In fact, CUPS is part of the 3GPP Release 14 standard because of the technology’s ability to distribute resources throughout the network, which is key to having an efficient 5G core network.
Some infrastructure vendors describe CUPS as another “tool in the toolbox” for 5G. “It’s a foundational concept as you move to 5G core,” said Martin McGrath, head of core technology for Nokia’s North America region. McGrath noted that while CUPS can be deployed in 4G networks, it is essential for 5G. “CUPS is a stepping stone to 5G,” he said.
Perhaps CUPS is getting so much attention now because infrastructure vendors are talking about the technology in relation to their 5G platforms. For example, ZTE earlier this month announced its 5G Common Core platform and the company noted that its platform incorporates CUPS, which makes it possible for networks to handle advanced tasks like network slicing.
Likewise, Affirmed Networks, which announced its 5G Mobile Core product last February, noted that its solution also uses a CUPS architecture.
McGrath said that CUPS is part of Nokia’s 5G Cloud Packet Core solution. And Ericsson has also incorporated CUPS into it virtual evolved packet core (vEPC) platform. Ericsson was recently ranked No. 1 in wireless packet core technology by Dell ‘Oro Group for the first quarter of 2018.
CUPS is essential to 5G networks because it allows operators to separate the evolved packet core (EPC) into a control plane that can sit in a centralized location, for example the middle of the country, and for the user plane to be placed closer to the application it is supporting. According to Angela Whitfield, vice president of marketing and product management at Affirmed Networks, this type of separation is key for applications such as the connected car. In that scenario, a network operator can place the EPC user plane in a data center in a city so that it is closer to the application and therefore reduces the latency. This scenario also works well for high-bandwidth applications like video. Because the core user plane is located closer to the end user the operator doesn’t have to backhaul traffic all the way to central hub and therefore saves money.
But Ericsson’s Staffan Lindholm, strategy product manager for the user plane at the vendor, noted that because many 5G network use cases require low latency, many operators are going to need to deploy more gateways to minimize latency created by transporting traffic.
Lindholm added that when an operator deploys a “dedicated” gateway for certain services, that’s considered a form of network slicing. “In the simplest form of network slicing you get dedicated gateways and you share a common core,” Lindholm said.
He added that, for example, an operator that deploys a 5G fixed wireless access service could be considered a network slice if it’s deployed using distributed gateways and a common core. Ericsson is one of Verizon’s 5G network vendors. The operator earlier this month launched its 5G fixed wireless service using pre-standard gear.
Nokia’s McGrath noted that while network slicing is part of many operators’ plans, most haven’t deployed it yet. Instead most are using their existing EPC. “When they deploy the 5G core, network slicing will be an inherent tool,” he added.