The problem of modeling and controlling resources in a system with interaction between hardware and software is considered. A model encompassing both hardware and software dynamics is developed together with an online estimation scheme in order reduce dependence on apriori information. A control structure is presented in order to control performance under constrained resource situations and to reduce effects of estimation errors and disturbances. The approach is applied to a conversational video case and evaluated through simulations.

A feedback-based scheme for cooperative content distribution for mobile systems aimed at reducing energy and licensed spectrum usage is presented. The paper is motivated by the problems in implementing peer-to-peer based content distribution in mobile networks due to the risks of unfair energy expenditure. A technique for incentivizing cooperation and for enforcing agreements between peers is discussed as well as the dynamics of the resulting barter-like economy. Rationale for a heuristic solver is provided together with simulation-based analysis of expected savings. The scheme is evaluated for cases with deterministic as well as random initial conditions.

No abstract

Abstract-The problem of resource management in a system of a-priori unknown software components executing on nondeterministic hardware is considered. The approach uses on-line parameter estimation to address uncertainties and combines this with a convex optimization-based control scheme able to handle overload situations. An algorithm to solve the optimization in real-time is presented together with performance analysis through simulations. An implementation of the approach is experimentally compared with a static analysis scheme using worst case a-priori estimates. It is demonstrated that the presented approach outperforms the static scheme in situations with uncertainty and that the advantage increases as uncertainty grows.

No abstract