Cloud-Edge Resource Scheduling and Offloading Optimization Based on Deep Reinforcement Learning.

As shown in the figure, except for the “No Offloading” strategy, all offloading methods exhibit a decreasing trend in both task drop rate and average latency as the edge node frequency increases from low to high, indicating that more tasks are successfully offloaded for processing, thereby alleviating the computational load on the terminal.

also in 47,180 other papers