DORA Workshop
High-Performance Heterogeneous Data Transfer for End–Edge–Cloud Robot Inference
Date, time, and room will be added once confirmed.
Talk overview
This session presents a unified data-transfer framework spanning CPU memory, GPU memory, local interconnects, and network links for robot inference. Using DORA's tensor-pool implementation, it covers persistent memory pools, topology-aware path selection, reusable receiver views, and resource-lifetime management.
