Function Dedicated Network Deployment Model in DPU/IPU-Like Device Acro...
summary
Claims that DPU/IPU cards plus an all‑photonic backbone can cut CPU use and reach sub‑10 ms XR motion‑to‑photon and net‑zero energy targets by 2030, using GPU Direct and RDMA to stream raw sensor/video into GPU/persistent memory and bypass host CPU. Current POCs run over PCIe Gen4 while waiting for Linux and CXL 3.0 support; vendors expose proprietary SDKs, so Open Programmable Infrastructure (OPI) aims to supply a common API for lifecycle and multi‑operator resource management.
AI-written summary · about
chapters (5)
- 0:00Start
- 2:02We are in the middle of the hardware evolution!
- 17:06Possibility for IGF Data Centric Infrastructure over APN
- 23:41Function Dedicated Network use case 1 alternative w/ wireless
- 26:02Function Dedicated Network use case 2
related talks
06mMaximizing Programmable DPUs in the Network
30mZXDH DPU Adapter and It's Application
19mCluster Optimization: Harnessing DPUs for Ambient Mesh Workloads
43mWhat is the Big Deal with the Open Programmable Infrastructure Project?
05mDay 1 Keynote
18mSmartNIC DPU Enablement in OpenStack with OVN
26mApplying New High-Throughput Network Solutions in IEC (Integrated Edge Cloud)
38mWhat Is the Big Deal with the Open Programmable Infrastruc...