develops the Distributed UPF with NVIDIA Bluefield-3 DPU
summary
NTT R&D builds a fully offloaded distributed UPF on Bluefield-3 DPU using DOCA flow APIs and the ConnectX‑7 eSwitch, running on 200 Gbps dual‑port cards; core claims are full data‑plane offload, carrier‑grade interoperability with DoCoMo 5GC on AWS (registration and session establishment), and measurable gains: 95% tail latency reduction and 13% power reduction versus host vUPF, plus a planned F5 POC for AI local breakout using N9 and BIG‑IP Next for Kubernetes.
AI-written summary · about
related talks
17mLightning Talk: eBPF based UPF (User Plane Function) for 5G
30mZXDH DPU Adapter and It's Application
22mIntegrating Cloud Native Data Plane (CNDP) with OMEC BESS-UPF
17mHiP4-UPF: Towards High-Performance Comprehensive 5G User Plane Function on P4...
19mCluster Optimization: Harnessing DPUs for Ambient Mesh Workloads
39mFunction Dedicated Network Deployment Model in DPU/IPU-Like Device Acro...
13mLightning Talk: Accelerating Cilium: Our Journey on Offloading Cilium To DPU
43mWhat is the Big Deal with the Open Programmable Infrastructure Project?
