Cloud resource distribution method based on variable neighborhood searching strategy

Abstract

The invention discloses a cloud resource distribution method based on a variable neighborhood searching strategy. A mode of renting according to demands is normally adopted by cloud computing resources, the demands of each service on the cloud resources are different for SOA applications formed by cloud services, and a calling relationship map among services is established to reasonably distributethe cloud resources to each cloud service; the service quality and deployment schemes of SOA mode cloud applications are defined; a multi-target cloud resource optimization distribution problem modelis established; a multi-target genetic algorithm based on the variable neighborhood searching strategy is adopted to solve a cloud resource optimized distribution problem. The invention provides thecloud resource distribution method based on the variable neighborhood searching strategy, the local searching capability of variable neighborhood searching and the global searching capability of the genetic algorithm are combined, the algorithm efficiency and the quality of resolutions are improved, and therefore the optimal cloud resource distribution problem that service performance is ensured and the cost is the lowest when the SOA applications are deployed is solved.

Description

technical field [0001] The invention belongs to the field of cloud computing resource management, and in particular relates to a cloud resource allocation method based on a variable neighborhood search strategy. Background technique [0002] With the popularity of cloud computing, more and more web application providers migrate their applications to cloud applications in a Service-Oriented Architecture (Service-Oriented Architecture, SOA) mode. Use the IaaS (Infrastructure as a Service) platform provided by the cloud provider to deploy your own distributed system, so that the application provider does not need to consider purchasing servers and related hardware equipment and maintenance, and only needs to rent the cloud provider as needed. provided cloud computing services. Therefore, it greatly reduces the threshold of Web system development and operation, shortens the development cycle, reduces maintenance costs, and makes operation and maintenance more convenient. It is ...

AI Technical Summary

Problems solved by technology

[0004] The optimal allocation problem has always been a research hotspot in the field of software engineering. However, the huge choice space and conflicting optimization goals make this problem a typical NP problem.

The current cloud resource optimization allocation methods assume that the number of deployed service instances and cloud resource instances is definite, or only consider the scaling characteristics of cloud resource instances, and do not consider the relationship between services and the scaling of service instances. It is difficult to generate an optimal solution that meets the demand and minimizes the cost

Images