TPAoI: Ensuring Fresh Service Status at the Network Edge in Compute-First Networking

TL;DR

TPAoI addresses the challenge of keeping service status fresh at the network edge by defining a three-phase AoI metric that captures forwarding, waiting, and requesting delays without heavy reliance on a single phase. The approach models the system as an MDP and solves it with a DRL method based on D3QN, using five sub-states to represent the three phases and their dynamics. Empirical results show TP AoI lowers AoI by about 47% compared with QAoI and reduces update frequency by about 48% versus conventional AoI under dynamic delays, demonstrating robustness to different delay distributions and user access patterns. This work offers a scalable, low-cost mechanism to maintain timely edge service status, with potential extensions to multiple edge servers and access points.

Abstract

In compute-first networking, maintaining fresh and accurate status information at the network edge is crucial for effective access to remote services. This process typically involves three phases: Status updating, user accessing, and user requesting. However, current studies on status effectiveness, such as Age of Information at Query (QAoI), do not comprehensively cover all these phases. Therefore, this paper introduces a novel metric, TPAoI, aimed at optimizing update decisions by measuring the freshness of service status. The stochastic nature of edge environments, characterized by unpredictable communication delays in updating, requesting, and user access times, poses a significant challenge when modeling. To address this, we model the problem as a Markov Decision Process (MDP) and employ a Dueling Double Deep Q-Network (D3QN) algorithm for optimization. Extensive experiments demonstrate that the proposed TPAoI metric effectively minimizes AoI, ensuring timely and reliable service updates in dynamic edge environments. Results indicate that TPAoI reduces AoI by an average of 47\% compared to QAoI metrics and decreases update frequency by an average of 48\% relative to conventional AoI metrics, showing significant improvement.

Figures (8)

• Figure 1: Illustration of compute-first networking
• Figure 2: AoI evolution in three-phase updating system
• Figure 3: Illustration of state $\mathcal{S}$w.r.t three phases
• Figure 4: Performance comparison of TPAoI, QAoI, and Con. AoI varying $\lambda$. (a) and (b) show the AoI comparison when transmitting delay fluctuation following $exp(1)$ and $\mathcal{N}(1,1)$, respectively. (c) and (d) show the corresponding total number of transmitted status updates from (a) and (b), respectively.
• Figure 5: AoI fitting $w.r.t$$\lambda$