Cloud-Network Resource Perception and Modeling Technology Based on Software-Defined Computing First Network

The Computing First Network (CFN) is a novel network paradigm that efficiently allocates and schedules computing force, storage resources, and network capacity across cloud, edge, and device domains. Leveraging the comprehensive network perspective offered by Software-Defined Networking (SDN) and precise assessment of distributed computing resources, this paper introduces an architecture known as Software-Defined Computing First Network (SD-CFN). This architecture forms the basis for modeling computing and network resources and designing a new representation of network topology. Ultimately, the proposed architecture and the novel network topology representation undergo comprehensive analysis and validation.

I. Introduction

With the evolution of deep learning, computing-intensive tasks like face recognition and target detection are on the rise. However, end devices might struggle to meet these demanding computational needs. This has led to the adoption of cloud computing ^{[1], [2]}. While cloud computing is adept at handling computational heavy lifting, it falls short when it comes to addressing delay-sensitive tasks. To tackle the latency issue, fog computing ^[1] and edge computing ^[3] have gained significant attention, offering potential solutions to mitigate the propagation delays associated with cloud computing.