Empowering Computing and Networks Convergence System with Distributed Cooperative Routing

TL;DR

This work addresses the challenge of scheduling computing and network resources end-to-end to meet strict deadlines in computing and networks convergence (CNC) systems. It proposes a distributed cooperative routing framework with four planes (trading, management, control, forwarding) and cross-plane cooperation to jointly optimize computation efficiency and network congestion. The key contributions include a detailed end-to-end routing pipeline with global system view maintenance, task splitting, and result feedback, plus simulations showing improved deadline satisfaction and lower costs compared with computation-first routing. The framework offers a scalable approach to tightly integrate computing and networking for latency-sensitive intelligent applications, with practical impact on CNC deployment and resource utilization.

Abstract

The emergence of intelligent applications and recent advances in the fields of computing and networks are driving the development of computing and networks convergence (CNC) system. However, existing researches failed to achieve comprehensive scheduling optimization of computing and network resources. This shortfall results in some requirements of computing requests unable to be guaranteed in an end-to-end service pattern, negatively impacting the development of CNC systems. In this article, we propose a distributed cooperative routing framework for the CNC system to ensure the deadline requirements and minimize the computation cost of requests. The framework includes trading plane, management plane, control plane and forwarding plane. The cross-plane cooperative end-to-end routing schemes consider both computation efficiency of heterogeneous servers and the network congestion degrees while making routing plan, thereby determining where to execute requests and corresponding routing paths. Simulations results substantiates the performance of our routing schemes in scheduling computing requests in the CNC system.

Figures (5)

• Figure 1: An illustration for convergence levels of CNC systems: resource convergence, function convergence and performance convergence.
• Figure 2: End-to-end computing request routing framework.
• Figure 3: End-to-end cross-plane cooperative routing schemes based on the computing request routing framework.
• Figure 4: A typical computing and networks convergence system.
• Figure 5: The completion cost of computing requests under different levels of network congestion. The lower the better. But when the cost equals to zero, it means the request cannot be routed under the deadline requirement.