A Reliable and Load Balancing Controller Placement Method in Software‐Defined Networks DOI Open Access

Mahsa Saeedi Goraghani,

Mahboubeh Afzali, Fazel Sharifi

et al.

International Journal of Communication Systems, Journal Year: 2024, Volume and Issue: 38(2)

Published: Dec. 17, 2024

ABSTRACT Software‐defined networking (SDN) achieves the programmability of control plane by separating it from data forwarding to provide flexible management network resources. The multicontroller architecture is required be deployed enhance scalability and reliability with traffic growth. However, controller placement problem (CPP) considered an important challenge in software‐defined networking, which should addressed. number controllers their locations are challenges that affect various aspects separated such as performance metrics, ability respond failures. Also, unappropriated subdomain partitioning multicontrollers may cause unbalanced distribution loads resulting reduction communication network. In this paper, optimization method based on particle swarm (PSO) algorithm presented for deploying CPP allocating switches controllers. proposed aims minimize cost known controllers, maximum load imbalance between improve resilience against a failure each switch its mapping controller. evaluated using two widely used networks Internet Topology Zoo Aarnet, Oxford, Chinanet, Interoute, ION topologies show method. results better propagation delay, balancing among when compared methods Varna, clustering‐based partition (CNPA), K ‐means. Moreover, improves CNPA, ‐means, respectively. PSO outperforms nearly 20% 17% decline comparison Varna‐based heuristic CNPA different scales topologies, enhances metric 6% case scenario Interoute shows improvement PSO, assigned 14%, 22%, 13%, 18% ‐means‐based ION, Aarnet topologies. Furthermore, 9%, 5%, 15% average delay Varna‐, CNPA‐, scheme higher failures existing approaches.

Language: Английский

A Reliable and Load Balancing Controller Placement Method in Software‐Defined Networks DOI Open Access

Mahsa Saeedi Goraghani,

Mahboubeh Afzali, Fazel Sharifi

et al.

International Journal of Communication Systems, Journal Year: 2024, Volume and Issue: 38(2)

Published: Dec. 17, 2024

ABSTRACT Software‐defined networking (SDN) achieves the programmability of control plane by separating it from data forwarding to provide flexible management network resources. The multicontroller architecture is required be deployed enhance scalability and reliability with traffic growth. However, controller placement problem (CPP) considered an important challenge in software‐defined networking, which should addressed. number controllers their locations are challenges that affect various aspects separated such as performance metrics, ability respond failures. Also, unappropriated subdomain partitioning multicontrollers may cause unbalanced distribution loads resulting reduction communication network. In this paper, optimization method based on particle swarm (PSO) algorithm presented for deploying CPP allocating switches controllers. proposed aims minimize cost known controllers, maximum load imbalance between improve resilience against a failure each switch its mapping controller. evaluated using two widely used networks Internet Topology Zoo Aarnet, Oxford, Chinanet, Interoute, ION topologies show method. results better propagation delay, balancing among when compared methods Varna, clustering‐based partition (CNPA), K ‐means. Moreover, improves CNPA, ‐means, respectively. PSO outperforms nearly 20% 17% decline comparison Varna‐based heuristic CNPA different scales topologies, enhances metric 6% case scenario Interoute shows improvement PSO, assigned 14%, 22%, 13%, 18% ‐means‐based ION, Aarnet topologies. Furthermore, 9%, 5%, 15% average delay Varna‐, CNPA‐, scheme higher failures existing approaches.

Language: Английский

Citations

1