Supporting Dependent Task at Edge: A Computation-Communication-Cache Integrated Service Architecture

In this paper, we propose a forwarding and processing mechanism based on the Information-Centric Networking (ICN) architecture that supports with dependent task requests. This mechanism enables multi-stage distributed decision-making, multi-node collaborative forwarding of dependent task requests, and distributed execution of dependent tasks at the edge. Furthermore, within the multi-stage distributed decision-making paradigm, we introduce an acceleration method based on parallelism between task decision-making and task execution. Additionally, the proposed mechanism incorporates an local state synchronization method for task decisions to reduce synchronization overhead. Finally, we build a prototype system to implement and evaluate the proposed forwarding and processing mechanism, verifying its functional correctness and demonstrating its performance advantages compared to both IP and ICN-based architectures.