Auction resource allocation mechanisms in grids of heterogeneous computers

This paper examines economic resource allocation through a number of auction types for a grid of e-waste computers. It examines the time to complete tasks and the energy usage of completing the tasks on a grid. A model of a simulated grid is developed and used to evaluate the resource allocation mechanisms. The model is an agent-based simulation where by user agents submit tasks to node agents that process these tasks. We evaluate three types of resource-allocator agents which all use a type of auction. The auction types are batch auction, continuous double auction and a pre-processed batch auction. The pre-processed batch auction is developed to try to have the advantages of both the continuous double auction and the batch auction. The simulated grid is calibrated to a real e-waste grid where each node has a performance index. This grid is a test grid of eight nodes of heterogenous computer hardware and with differing computational ability and energy usage. We simulate the auction types under the same task input streams. We consider a task impulse response stream on energy usage and time to complete all tasks and a input stream step response. Finally we consider the three auction allocation mechanisms under a random task stream. The paper finds that the choice of auction method makes a substantial difference in the time to complete tasks and in total energy consumption