Service providers are investing billions of dollars in their network infrastructure to keep up with demand for a wide range of new services as 5G Networks usage increases quickly. Choosing where to distribute compute resources in their networks to meet the specific performance requirements of these varied applications is a key challenge in handling the fast expansion of 5G.
The core and edge data centers of service providers must be able to provide the 5G compute capacity requirements for ultra-low latency and high throughput services. By effectively managing heavy data usage apps, new developments in the design of the 5G user plane function (UPF) can help them improve their packet processing capabilities and meet demand.
The user plane and control plane of the network may now be separated for incredibly effective operation, making the UPF the workhorse for the flow of digital data traffic. However, if too many packets pass through the network at once, the UPF risk becoming a bottleneck. It takes a fresh strategy using cutting-edge software technology to get rid of this.
To reduce the additional load on the user plane and achieve a performance of many terabytes per second, larger capacity hardware has generally been used. Although this strategy may handle rising demand, it cannot effectively handle the complexity of related software and protocols. Service providers won’t be able to guarantee a consistent performance if they rely entirely on data processing hardware as a rising number of edge applications demand the placement of UPFs closer to customer locations.
Red Hat, Intel, and Kaloom worked together to create an edge-based UPF that can use a hybrid hardware acceleration approach to give over 4 million subscribers with 2.2 Tbps (full duplex) throughput. With Kaloom’s Unified Edge Fabric™, Red Hat’s OpenShift™ on P4-programmable switches, based on Intel® Tofino™ Intelligent Fabric processors, Intel’s Stratix-10 FPGAs, and Intel Xeon processors, the technical cooperation between the three organizations exemplifies the most recent advancements in UPF packet processing.
Want to know more about Red hat? Visit our course now.
With this new concept, network latency is considerably reduced from milliseconds to microseconds while performance is improved at a lower cost. This enables service providers to satisfy the requirements of 5G-enabled applications like the internet of things (IoT), augmented and virtual reality (AR/VR), and many more.
To meet changing demands for increased bandwidth, lower latency, and subscriber scalability, service providers can develop more quickly and introduce new services using an innovative UPF software design that makes use of heterogeneous programmable hardware. Red Hat OpenShift offers a platform for building cloud-native applications and installs the entire system with a small footprint. In comparison to traditional server installations, Intel® Tofino™ Intelligent Fabric processors, Intel’s Stratix-10 FPGAs, and Intel Xeon processors provide better throughput, lower latency, and an improved cost-performance ratio.