|
| Ref | Policy/algorithm/strategy | Problem addressed | Improvement/achievement | Weakness/limitation |
|
| [129] (2021) | Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) and MCDM algorithm | Network performance-based blocking probability, jitter, delay metrics, and assignment of resources problems and challenges inside the controller | Improved network resource allocation path. Improved blocking probability. Identifies the path that best suits each flow | Limited resource utilisation. Needs further improvement with machine learning algorithms. The proposed algorithm is not dynamic and flexible |
| [122] (2020) | Naive-ESN-based services awareness algorithm | Challenges and demands presented by OpenFlow. Controlled software-defined optical networks (SDONs) and the issue of matching between services and SDON | Improved the performance of several services in SDON | Limited to only ring topologies |
| [54] (2019) | OLA-based multicast optimization scheme | The major challenge is the growing market for multidiversion services, and the size of the software-defined optical networks (SDON) is significantly affected by optical layer influence | The paper suggested a special approach for software-defined optical networks based on optical layer concepts. This method created a paradigm of optical layer sensitivity for the optical signal | More room for improvement in bit error rate, efficiency, and time setup performance |
| [130] (2019) | ROADM and physically centralized architecture (PCA) | Difficulty in managing network dynamism in switching, routing, and middle box component band service providers | Adequate power usage and less expensive for small- or medium-sized network structures that have a centralized controller | Quality factor, signal-to-noise ratio, bit error rate, and overall performance are limited to 45 km |
| [56] (2019) | Duplication of SDN controller based on VNF | For balancing of network load for easier distribution of resources | Improvement in the “utilisation of bandwidth” and the overall throughput reducing the delays | More complex and cost ineffective since multiple SDN controllers are to be used. High power and energy usage |
| [57] (2019) | “Tenant” allocation of network resources | To manage the throughput and bandwidth efficiently | Improved throughput of the network and well-met traffic standards | High latency and unreliable approach |
| [58] (2017) | Bayesian classifier | Less service consciousness when operating along with an OpenFlow protocol and less efficiency in bandwidth scheduling. These service areas in SDON need improvement | This paper proposes the use of the platform “Bayesian classifier-based service-aware (BC-SA).” The BC-SA system has been implemented using the Bayes network paradigm for multiple services. This BC-SA technique was used by the SDON system to carry out more efficient planning of resources | SDON is unable to cope with big data and a wide range of characteristics and demands |
|
