.png){kind=icon}
.png){kind=icon}
Return
Multi-objective optimization algorithm for VNF migration with priority awareness in dynamic networks
.png){kind=icon}
PRE
AI
DOI：10.1016/j.comnet.2025.111666.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
With the continuous development of Network Function Virtualization (NFV) technology, Virtual Network Function (VNF) migration has become a crucial approach to optimizing network resource utilization, reducing service latency, and improving service quality. However, in dynamic network environments, VNF migration faces challenges such as resource overload, service request prioritization, migration cost optimization, routing overhead, and energy consumption. To address these challenges, this paper proposes a priority-aware and multi-objective optimization-based VNF migration algorithm, namely the Lagrangian Fish Optimization for VNF Migration (LFO-VNM) Algorithm. This algorithm integrates the Lagrangian relaxation method with the Artificial Fish Swarm Algorithm (AFSA) to dynamically adjust resource allocation and migration paths, optimizing migration cost, network performance, and node energy consumption while prioritizing high-priority service requests.First, a Mixed-Integer Linear Programming (MILP) model is established to quantify the impact of VNF migration on network link load, node resource consumption, and service performance. Based on this, a multi-objective optimization model is formulated, considering network bandwidth, latency, migration cost, and energy consumption. This model is decomposed into a series of linear subproblems, which are more efficiently solved using the Lagrangian relaxation method. Finally, leveraging the global search capability of AFSA, an efficient solution algorithm, LFO-VNM, is designed to optimize VNF migration decisions. Experimental results demonstrate that the proposed algorithm not only improves computational efficiency but also effectively reduces total cost and energy consumption, outperforming existing migration algorithms across various network topologies. This study provides an effective solution for VNF migration and resource scheduling in complex network environments.
Journal
IF:
4.6
Papers:
1.5K
Citations:
1.6W