XDP Offload for OPC UA
summary
Experimental setup uses XDP sockets merged into the mainline kernel to deliver zero copy packets into userspace where a lightweight LWIP network stack runs, bypassing Linux kernel network processing for OPC UA receive-path logic. Implementation requires a patched driver and careful tuning of socket/CPU scheduling; results show measurable latency reductions for request-to-measurement reads but gains vary by workload and hardware — in some tests only around one percent.
AI-written summary · about
chapters (7)
- 0:00Introduction to XDP Offload
- 3:30OPC UA protocol basics
- 7:55Linux networking and performance
- 12:47Callback latency in applications
- 17:53Optimizing XDP for OPC UA
- 22:26Compatibility and optimization tips
- 25:19Q&A and summary
related talks
28mLow-Latency, Deterministic Networking with Standard Linux using XDP Sockets
18mOpen vSwitch with AF_XDP as a userspace datapath
30mZXDH DPU Adapter and It's Application
19mGet start with XDP: write a network filter in 10m
22mKernel-Native Packet Processing on AMD GPUs: XDP Offload without Userspace Runtimes
33mBringing packet queueing to XDP
21mRemote TCP Connection Offload and Applications
16mElectrode: Accelerating Distributed Protocols with eBPF