Federated Cloud computing as System of Systems

Energy and resource utilization efficiency are two of the most significant differentiators in the contemporary Cloud computing marketplace. This paper addresses both of these from system of systems architectural and optimization viewpoints. Our approach will allow multiple cloud providers to optimally utilize compute resources. It will do this by (1) lowering the datacenters deployments per provider ratio, share and (2) scheduling available energy via aggregators and (3) lastly to employ, where appropriate, more renewable and carbon free engines. This is an embodiment of a cloud-federation, and will be treated as a System of Systems. This new cloud federation SoS paradigm will develop a new control methodology that is both financially attractive and energetically and environmentally advantageous. It supports dynamic expansion and contraction of computing capabilities for handling sudden variations in service demand. As is expected, the constituent cloud services retain independent ownership, objectives, funding, and sustainability means. While individual system behavior within the SoS is independently managed, the SoS as control structure might introduce cascading effect that impacts the overall SoS behavior. This paper analyzes the core SoS requirements, concept synthesis, and functional architecture. Finally, we suggest a physical architecture that simulates some of the main SoS emergent behavior to diminish unwanted outcomes while encouraging desirable results.