Energy-aware cloud workflow scheduling optimization method based on multi-population genetic algorithm

Abstract

The invention discloses an energy consumption perception cloud workflow scheduling optimization method based on a multi-swarm genetic algorithm, comprising the following steps: obtaining information required for scheduling optimization; calculating task level values; initializing populations based on levels; Calculate the fitness value; communicate between subpopulations; each subpopulation evolves independently: perform crossover and mutation operations based on two-dimensional topological sorting to form a new subpopulation, use the FBI&D method to improve the new subpopulation and calculate the fitness value. The new sub-population forms a new contemporary sub-population; until the termination condition is satisfied, the scheduling optimization scheme is output. The invention considers energy consumption factors and adopts multi-group co-evolution strategy, which can effectively prevent the population from entering local optimum and premature maturity, and accelerate the convergence, thereby improving the efficiency of the entire algorithm.

Description

technical field [0001] The present invention relates to the fields of computer technology, information technology and system engineering, in particular to a cloud workflow scheduling optimization method, more specifically, to an energy-aware cloud workflow scheduling optimization method based on a multi-population genetic algorithm. Background technique [0002] Workflow under the cloud computing environment, referred to as "cloud workflow", is the integration of cloud computing and workflow-related technologies, and has a wide range of applications in cross-organizational business collaboration and scientific computing that require efficient computing performance and large-scale storage support. prospect. In cloud workflow, there are timing constraints between tasks, and virtual machines are usually used as the smallest allocation unit of computing resources to receive and process these tasks during execution. Cloud workflow scheduling optimization refers to how to allocat...

AI Technical Summary

Problems solved by technology

[0012] In order to overcome the quality of the heuristic method solution is usually not very high and depends on the type of workflow, combined with the heuristic semi-intelligent computing method, the intelligent computing method based on layered coding and the incompleteness of the coding search space, the one-dimensional coding based Due to the large number of many-to-one relationships between individuals and scheduling schemes, the existence of a large number of redundant coding search spaces, and the use of global search, the search efficiency will be reduced, and energy consumption factors are rarely considered in the intelligent computing method. The present invention provides a Energy-aware cloud workflow scheduling optimization method based on multi-population genetic algorithm, which effectively improves the efficiency and quality of the solution

Images