RIPPLE: Lifecycle-aware Embedding of Service Function Chains in Multi-access Edge Computing

TL;DR

RIPPLE tackles lifecycle-induced delays in MEC-based SFC embedding by jointly considering user connectivity forecasting and VNF lifecycle dynamics. It decomposes the problem into forecasting, lifecycle-aware embedding, and link embedding, deploying a heuristic that pre-activates VNFs at locations with high handover probability to meet latency requirements. Empirical results on tree and city topologies show RIPPLE substantially reduces service interruptions, closely approaching an ideal instantaneous-lifecycle benchmark and outperforming reactive baselines. The work demonstrates the practical importance of lifecycle awareness for robust SFC management in dynamic MEC environments.

Abstract

In Multi-access Edge Computing networks, services can be deployed on nearby edge clouds (EC) as service function chains (SFCs) to meet strict quality of service (QoS) requirements. As users move, frequent SFC reconfigurations are required, but these are non-trivial: SFCs can serve users only when all required virtual network functions (VNFs) are available, and VNFs undergo time-consuming lifecycle operations before becoming operational. We show that ignoring lifecycle dynamics oversimplifies deployment, jeopardizes QoS, and must be avoided in practical SFC management. To address this, forecasts of user connectivity can be leveraged to proactively deploy VNFs and reconfigure SFCs. But forecasts are inherently imperfect, requiring lifecycle and connectivity uncertainty to be jointly considered. We present RIPPLE, a lifecycle-aware SFC embedding approach to deploy VNFs at the right time and location, reducing service interruptions. We show that RIPPLE closes the gap with solutions that unrealistically assume instantaneous lifecycle, even under realistic lifecycle constraints.

Figures (8)

• Figure 1: Burst length with respect to an increasing horizon, tree topology. Inference chain trained on $\alpha=0.9$.
• Figure 2: CDF of users with a certain ratio of unsuccessful packets. Inference chain trained on $\alpha=0.9$.
• Figure 3: Number of VNFs prepared with respect to different mobility correlation (see legend). Inference chain trained on $\alpha=0.9$.
• Figure 4: CDF of users with a certain ratio of unsuccessful packets. Inference chain trained with $\alpha=0.5$.
• Figure 5: Number of VNFs prepared with respect to different mobility correlation (see legend). Inference chain trained on $\alpha=0.5$.