It is quite common for both parents in a family to work. Sometimes this is because they need two salaries to survive whereas other times it is their choice. What are the problems you see in this situation? How does this affect the children and the family life?

The main objective of applying self-structuring artificial intelligence for Software Defined Networks (SDN) in

Data

Centres is to minimise the human intervention on managing SDNs effectively to optimise the

network

performance, service orchestration, troubleshooting and proactive fault mitigation. Nowadays,

data

centres are an essential resource due to the progressive digitalization of organizations and the technologically advancing world. Modern

data

centres are driven by SDNs to facilitate the dynamically changing traffic demand and modern

network

application requirements

such

as high volatile resource provisioning which needs flexible and agile control of the

Data

Center

network

infrastructure. The resources within an SDN are becoming more sophisticated and heterogeneous,

such

that it is challenging to effectively organise, manage, secure and optimise the SDN manually. Even though conventional Artificial intelligence (AI) algorithms have been applied to overcome some of these challenges,

however

, the need for a human operator still exists in either the design, application or deployment of

such

algorithms and models.

This

research aims to apply Self-Structuring Artificial Intelligence (SS-AI), a new type of AI that continuously adapts to the environment, to SDNs which would mitigate difficulties of pre-determining a suitable AI architecture in conventional AI for more complex, non-deterministic and dynamic environments like

data

centre

networks. Metrics and

data

about the

Data

Centre

network

will be captured, transformed, and streamed in real-time(near real-time) into the AI layer using a

data

streaming platform. Different SS-AI models could be created to represent different domains which could be used to detect anomalies, patterns, trends and other useful features.

Moreover

, these models can be fused together to

further

identify correlations and relationships.

This

research would produce a minimal viable product (MVP) that could be deployed in a real-world

data

centre

which have features

such

as comprehensive visualization of the operational metrics of the

data

centre

, proactive fault detection and root cause analysis, resource requirement prediction based on historical trends, real-time (near real-time) automated and optimal

network

management and service orchestration.