Defense Strategies for Multi-Site Cloud Computing Server Infrastructures

We consider cloud computing server infrastructures for big data applications, which consist of multiple server sites connected over a wide-area network. These sites house a number of servers, network elements and local-area connections, and the wide-area network plays a critical, asymmetric role of providing the vital connectivity between them. We model this infrastructure as a system of systems, wherein the sites and wide-area network are represented by their cyber and physical components. These components can be disabled by cyber and physical attacks, and also can be protected against them using component reinforcements. The effects of attacks propagate within the individual systems, and also beyond them via the wide-area network. We characterize these effects using correlations at two levels using: (a) aggregate failure correlation function that specifies the infrastructure failure probability given the failure of an individual site or network, and (b) first-order differential conditions on system survival probabilities that characterize the component-level correlations within individual systems. We formulate a game between an attacker and a provider using utility functions composed of survival probability and cost terms. At Nash Equilibrium, we derive expressions for the expected capacity of the infrastructure given by the number of operational servers connected to the network, using the sum-form, product-form and composite utility functions.