- Tytuł:
- A hybrid evolutionary algorithm of optimized controller placement in SDN environment
- Autorzy:
-
Hemagowri, J.
Tamil Selvan, P. - Powiązania:
- https://bibliotekanauki.pl/articles/38704829.pdf
- Data publikacji:
- 2023
- Wydawca:
- Instytut Podstawowych Problemów Techniki PAN
- Tematy:
-
controller
software defined network
Gaussian chaotic map
fish swarm
multi-criteria optimization
kontroler
sieć zdefiniowana programowo
mapa chaosu Gaussa
rój ryb
optymalizacja wielokryterialna - Opis:
- Controller placement problem (CPP) is a significant technological challenge in software defined network (SDN). Deployment of a properly designed SDN-based network is required to detect optimal number of controllers for enhancing the network’s performance. However, the best possible controller placement for enhancing the network’s performance faces many issues. To solve the CPP, a novel technique called the hybrid evolutionary algorithm of optimized controller placement (HEA-OCP) in SDN environment is introduced to increase network’s performance by different network topologies. In the proposed model, optimized controller placement using improved multi-objective artificial fish optimization is employed to improve data transmission and reduce latency. Controller placement can be determined using an undirected graph based on a variety of factors, including propagation delay, load balancing capabilities and bandwidth, fault tolerance and data transfer rate, and a variety of other factors. For each controller, the fitness value is calculated over multi-criteria functions. The optimizer’s performance can be improved with the use of Gaussian chaotic maps. In large-scale SDN networks using HEC-OCP, the algorithm dynamically analyzes the optimal number of controllers and the best connections between switches and controllers. As a result, the overall network performance is improved and the delay minimization-based controller placement strategy is obtained. The simulation of HEA-OCP with existing methods is conducted by a network topology dataset of various metrics, namely packet delivery ratio, packet drop rate, throughput, average latency, and jitter. The proposed HEA-OCP improves the packet delivery and throughput with reduced average latency, and packet drop ensures more instantaneous communications in real-time applications of SDN for better decision-making.
- Źródło:
-
Computer Assisted Methods in Engineering and Science; 2023, 30, 4; 539-556
2299-3649 - Pojawia się w:
- Computer Assisted Methods in Engineering and Science
- Dostawca treści:
- Biblioteka Nauki