Distributed resource allocation method based on inframarginal analysis and game

Abstract

The invention relates to the field of distributed technology, in particular to a resource allocation method based on inframarginal analysis and game theory in new classical economic theory. The method provided by the invention comprises the following steps: determining the comparative advantage nodes of a task; judging the state of the node sequence; and adopting a game inframarginal analysis mechanism for the node sequence conforming to the conditions, wherein the flow of the game inframarginal analysis mechanism is as follows: inputting a comparative advantage task matrix T of the node, and inputting a network condition matrix K of the node; determining the node sequence for performing inframarginal analysis; and judging the allocation mode selected among nodes according to the corresponding value of the node sequence in K. The method provided by the invention determines the corner solution through inframarginal analysis, shortens the solution space, controls the phenomenon of space explosion, and introduces the new classical economic theory and enables the node to become the unity of resource provider and user so as to solve the problem that some states cannot be analyzed as the resource provider and user are separated originally; and finally, the method realizes Pareto optimality of the system, namely maximum global utility, and achieves the aim of maximum system utility.

Description

technical field [0001] The invention relates to the field of distributable technology, in particular to a resource allocation method based on ultramarginal analysis and game theory in emerging classical economic theory. Background technique [0002] With the emergence of massive data, distributed computing is more and more widely used in theoretical computing and various applications. The idea of ​​distributed computing is to divide the project data that requires a lot of calculations into small pieces, and calculate them separately by multiple computers, and then upload the calculation results and combine them to draw data conclusions. There are many technologies developed accordingly, such as grid computing, voluntary computing and cloud computing, each of which has its own characteristics. [0003] 1. Grid Computing proposed by Foster et al. is the most mature distributed computing in current applications. Grid computing organizes computers scattered in different geogra...

AI Technical Summary

Problems solved by technology

The data in a distributed environment is massive. Although it is effective to study the maximization of income in limited resources, there is an upper limit to the income. More importantly, there is no import for technological progress.

Images