A low power scheduler using game theory

Abstract

In this paper, we describe a new methodology based on game theory for minimizing the average power of a circuit during scheduling in behavioral synthesis. The problem of scheduling in data-path synthesis is formulated as an auction based non-cooperative finite game, for which solutions are developed based on the Nash equilibrium function. Each operation in the data-path is modeled as a player bidding for executing an operation in the given control cycle, with the estimated power consumption as the bid. Also, a combined scheduling and binding algorithm is developed using a similar approach in which the two tasks are modeled together such that the Nash equilibrium function needs to be applied only once to accomplish both the scheduling and binding tasks together. The combined algorithm yields further power reduction due to additional savings during binding. The proposed algorithms yield better power reduction than ILP-based methods with comparable run times and no increase in area overhead.

Energy‐aware virtual machine consolidation based on evolutionary game theory

“…Game theory is a tool for analyzing the interaction of decision makers with conflicting objectives 34 . It is concerned with situations in which decision‐makers interact with each other, and in which each participant's satisfaction with the outcome depends not just on his own decisions but on other people's decisions 35 .…”

“…Each player can make decisions individually, therefore, game theory can be used to design and analyze distributed system. The features of game theory are 34 : (i) Selfish: each player is always selfishly trying to optimize its gain. (ii) Coalition: coalition formation is significant when a subset of players has the same agenda in terms of strategies for optimization.…”

A Multi-Objective Perspective for Operator Scheduling Using Finegrained DVS Architectures

“…Several heuristic methods, such as those based on genetic algorithm [15] produce near-optimal results in a reasonable CPU time. A game theory-based ap-proach is presented in [17], which has considerable CPU running time for large VLSI systems. Traditionally resource and latency has been the most important optimization objectives.…”