Sprint unveils C3PO for open source NFV/SDN

C3PO is part of the ever-expanding toolbox of solutions to help Sprint meet the coming wave of data in the years ahead.

Wait no more: After four years of research and development with Intel Labs, Sprint today unveiled the open source effort it calls C3PO, short for the clunkier Clean CUPS Core for Packet Optimization, where CUPS stands for Control & User Plane Separation.

C3PO is an open source NFV/SDN-based mobile core reference solution designed to significantly improve performance of the network core by providing a “clean, streamlined, high-performance data plane for the packet core,” Sprint said.

“C3PO revolutionizes the network core and it’s part of our expanded toolbox of solutions to meet the coming wave of data in the years ahead,” said Günther Ottendorfer, COO for technology at Sprint, in a press release. “C3PO is an important part of our NFV and SDN initiative, enabling Sprint to adapt more quickly to market demands and scale new services more efficiently and cost-effectively.”

C3PO uses standard high-volume server hardware and streamlines mobile core architecture by collapsing multiple components into as few network nodes as possible. In fact, by combining Sprint’s real-world operator knowledge with Intel’s research on optimizing software for standard high-volume servers, they developed a single solution that provides seven functions previously located within separate physical elements, according to Ron Marquardt, vice president of technology at Sprint.

In lab tests conducted on Dell EMC DSS 9000 rack scale infrastructure with compute sleds running dual socket 14 core Intel Xeon processors E5-2680 v4 with hyperthreading disabled, Sprint achieved a benchmarked forwarding rate of 1.63 million packets per second (Mpps), Marquardt said in a blog post.

The C3PO configuration demonstrated high efficiency with a single packet processing Intel Xeon processor core, performing a total of seven virtualized packet functions necessary for all network traffic for 500,000 subscribers using a typical Sprint traffic model. In addition to the packet processing core, six other Intel Xeon cores are used (Rx core, Tx core, Load balancing core, Interface core, Master core and Control Plane core), the company said. A C3PO configuration with one packet processing core achieved 2.2 Mpps for 50,000 subscribers with a similar traffic model. ng4T RAN emulators and MME were used as the test harness for benchmarking.  

Marquardt explained that over the past few years while preparing to move from proprietary hardware to virtualized solutions, it became increasingly important for Sprint to understand how to best optimize its network applications for provisioning on an Intel processor-based server.  

“When our work with Intel Labs developing C3PO first began, CUPS was just a nascent concept, and the industry work being done on NFV was still theoretical,” he said. “For simplicity, the latter was often depicted as individual software elements conceptually interconnected in a manner that would require significant compute overhead. NFV and SDN were both defined in their own contexts, with little attempt yet to marry these concepts in a performance-optimized manner. It was the early days of the industry’s march to virtual implementations.”
Sprint set out to find a more efficient solution. “As a result, we identified seven network functions applicable to essentially all user traffic, serving gateway, packet gateway, deep packet inspection, child protection filtering, carrier-grade NAT, static firewall, and service function chaining classification,” Marquardt said. “Instead of connecting these as separate functions, we developed C3PO as a way to collapse the functionality together into one implementation for the network core.”

Sprint's schematic of the C3PO architecture.

C3PO is designed to be used by global operators and other third parties as a reference for commercial applications. Sprint’s role with C3PO was to write the SDN piece, which it has now contributed to OpenDaylight. Intel Labs developed the Evolved Packet Core control and user plane and optimized them for basic packet forwarding and the seven network functions. Intel Labs contributed its code to the CORD project via the Linux Foundation and ON.Lab.

RELATED: Sprint joins CORD Project focused on SDN, NFV, cloud

This past April, Sprint became the latest U.S. operator to join the CORD Project, the community working to advance the Central Office Re-architected as a Datacenter (CORD) open source delivery platform that combines SDN, NFV and cloud technologies.

Sprint is not a member of the Open Network Automation Platform (ONAP) project that resulted from the merger of open source ECOMP and Open-O. A Sprint spokesperson said it is following ONAP carefully but focusing its efforts with C3PO on its portion of the project.