Market-based model predictive control for survivable distributed information systems: Resource allocation and algorithm selection

As modern networks can be easily exposed to various adverse events such as malicious attacks and accidental failures, there is a need to study their survivability. There are several important trends of modern information networks. They tend to be large-scale with distributed and component-based architectures, and the dynamic nature of operating environments leads them to utilize alternative algorithms. As a result, the behavior of such an information network can be controlled through resource allocation as well as algorithm selection. We study an information network that characterizes such trends. The service provided by the network is to produce a global solution to a given problem, which is an aggregate solution of partial solutions of individual tasks. Quality of service of the network is determined by the value of global solution and the time taken for generating global solution. In this thesis we design a scalable adaptive control mechanism along the lines of model predictive control to support the survivability of such networks by utilizing resource allocation and algorithm selection. To address adaptivity we model stress environment by quantifying resource availability through sensors. We build a mathematical programming model with the resource availability incorporated, which invokes optimal control actions as a function of both state and stress environment. The programming model is then decentralized through an auction market. By periodically opening the auction market, the system can achieve desirable performance adaptive to changing stress environment while assuring scalability property. We verify the designed control mechanism empirically.