Copenhagen, Denmark – Napatech, the provider of programmable smart network interface cards (SmartNICs) used for data processing unit (DPU) and infrastructure processing unit (IPU) services in telecom, cloud, enterprise, cybersecurity and financial applications worldwide, announced the delivery of the release of its SmartNIC-based offload solution for 5G User Plane Function (UPF), enabling hardware and software ecosystem partners to start integrations with their products.
Unlike most 4G networks based on purpose-built appliances, the 5G packet core is implemented as virtualised or cloud-native software running on servers located within edge and core data centres. As communications service providers (CSPs) and enterprises scale up the deployments of their 5G networks, they face strong financial pressure to maximise the number of users that can be supported on each server, whether individual subscribers or IoT devices, thereby minimising the net cost-per-user.
Within 5G packet core software, the subsystem that represents the highest compute workload is the User Plane Function (UPF), which performs critical packet inspection, routing and forwarding functions. Since general-purpose server CPUs are not well suited to the performance and latency requirements of real-time packet processing, network operators and 5G core software vendors increasingly adopt solutions for offloading the UPF to accelerator cards optimised for executing such workloads.
Napatech addresses the key business challenges around packet core deployments through its integrated hardware/software solution that comprises a fully-offloaded UPF fast path implemented within Napatech’s Link-Inline software stack, running on an FPGA-based programmable SmartNIC. The UPF data path is implemented as a port-to-port inline or “hairpinned” architecture, which ensures that following initial setup all flows are processed on the SmartNIC with no need to pass traffic to and from the server CPU, maximising the overall performance of the system. In line with plans announced earlier this year, the first version of this solution is now being delivered to selected hardware and software ecosystem partners.
Using a single 200Gbps SmartNIC to sustain 100Gbps of full duplex traffic, the Napatech UPF offload solution processes up to 140 million concurrent flows, with a flow learning rate greater than 1.5 million flows per second, achieving a total throughput of up to 85 million packets per second on stateful connections and ensuring full wire speed operation for typical packet sizes. In a representative use case analysed by the company, the Napatech UPF offload solution enables network operators to support 75x more users per server than with a software-based UPF and 7x more users per server than with a competing ASIC-based SmartNIC.
“Through collaboration with our partner ecosystem, Napatech ensures that our UPF offload solution is available to CSP and enterprise end-users through their preferred software vendors, server OEMs and system integrators”, says Charlie Ashton, senior director of business development, Napatech. “We are now delivering the first version of this solution to selected hardware and software partners, which we anticipate will lead to the availability of integrated, system-level solutions ready for deployment in the first half of 2023.”
“AvidThink was pleased to feature Napatech’s 5G UPF offload solution in our 2022 SmartNICs and infrastructure acceleration report published in April, and we are delighted to see them delivering the first release in this second half of the year,” says Roy Chua, founder and principal, AvidThink. “As CSPs and enterprises look to deploy 5G packet cores in edge clouds and on-premises data centres, Napatech’s solution can enable these deployments while reducing energy consumption, cost and footprint, contributing significantly to improving the ROI for their network infrastructure.”
Napatech will showcase its 5G UPF offload solution at Layer123 World Congress 2022 in London from Dec. 5th through 7th.
For more information on Napatech’s 5G UPF offload solution please see the solution brief.
Comment on this article below or via Twitter: @VanillaPlus OR @jcvplus”
