A Batch Scientific Workflow Task Scheduling Algorithm Based on Improved Genetic Algorithm

Abstract

The invention relates to a batch-processing scientific workflow task scheduling algorithm based on an improved genetic algorithm, comprising the following steps: generating an initial population according to an integer encoding method; continuously performing cross-mutation operations of a genetic algorithm according to a fitness function value, and eliminating inferior solutions , to generate a new high-quality solution; after multiple iterations, an optimal solution is retained according to the fitness function; on the basis of not changing the critical path of the optimal solution, shrinking the number of virtual machines used in the non-critical path is performed to obtain the final solution, and the final solution output. The invention has a reasonable design, can efficiently perform task scheduling of batch scientific workflow, is beneficial to reduce task scheduling cost and task scheduling time generated in the process of batch scientific workflow task scheduling, and can be widely used in various scales Cost-oriented cloud computing task scheduling.

Description

technical field [0001] The invention belongs to the technical field of cloud computing task scheduling, in particular to a batch processing scientific workflow task scheduling algorithm based on an improved genetic algorithm. Background technique [0002] With the development of computer network technology, technology has entered the era of big data, and a large amount of data is generated every moment. Data is the driving force behind the development of new technologies, and it contains endless business opportunities, making more and more enterprises invest in the research of the data field, and cloud computing technology emerged as the development of distributed and grid computing. In the era of big data, how to effectively manage the increasing amount of data and how to extract useful information from a large amount of data, that is to say, the transformation of data from "number" to "data", has become a problem for IT companies and Research institutions urgently need to...

AI Technical Summary

Problems solved by technology

Although many scheduling algorithms for general workflows can be used for batch workflows, they tend to lead to low resource utilization because they do not consider the batch structure characteristics of batch workflows.

Cai et al. proposed a hybrid heuristic algorithm with the least new lease time interval priority rule, the cheapest execution cost priority rule, and the remaining time slice length and execution time matching priority rule, but the algorithm defaults between task batches and task batches. Independent of each other and the randomness of the initial resource allocation scheme is too large, it is easy to cause the algorithm to converge too slowly, and the search for the optimal solution mainly depends on the virtual machine upgrade, and less attention is paid to the degree of parallelism

[0004] It can be seen that the above classic algorithms have many limitations, the task scheduling results are not ideal, and the complexity is high, it is difficult to meet the requirements of actual batch scientific workflow task scheduling

In summary, the existing batch scientific workflow task scheduling algorithms have a lot of room for improvement in terms of task scheduling cost and task scheduling time

Images