Font Size: a A A

Research On Cross-platform Cloud Service Portfolio Recommendation And Integration Methods

Posted on:2019-06-02Degree:MasterType:Thesis
Country:ChinaCandidate:B WangFull Text:PDF
GTID:2428330566989021Subject:Engineering
Abstract/Summary:PDF Full Text Request
With the introduction of more and more cloud computing companies and different functional cloud services,the application of cloud services has entered the era of universalization,bringing great convenience to users.However,due to the lack of unified standards among cloud platforms,different interface specifications and data formats make it difficult for developers to migrate applications between different platforms.In addition,due to the different technical characteristics of each platform,the cloud services introduced by it have different features and cost-effectiveness.This also leads to the difficulty for users to maximize their profits on a single platform.Therefore,how to implement multi-platform background Providing users with the optimal service portfolio and integrating the composite cloud services to ensure the compatibility of the combination and the stability of the performance have become an urgent problem to be solved.On the basis of fully analyzing the current mainstream cloud computing products,this paper makes a detailed classification of cloud services from different perspectives of functions and performance,and introduces semantic technologies to build a multidimensional semantic ontology model of cloud services to realize cloud A unified description of the service.At the same time,it fully considers the relationship and dependencies of cloud services at different levels and different attributes.It uses relevance data method to construct the semantic association network of cloud services,and constructs the corresponding semantic inference rules to realize the correlation service inference.In addition,the paper also sets weights for the relationship between cloud services and the credibility of reasoning,and proposes the concept of association degree and hybrid cloud service recommendation algorithm to recommend the optimal cloud service composition for users.In addition,this article also designs and develops a cross-platform cloud service integration system based on interface encapsulation,which is aimed at REST-style cloud services.An interface encapsulation method based on JSON description is proposed.Thepurpose of cross-platform cloud service integration is achieved by automatically generating SDKs containing service invocation interfaces for different platform cloud services.This method has the characteristics of loose coupling,fast integration and convenient integration between services.It solves the problem that cross-platform cloud service calls require multiple authentication and manual connection problems,which greatly reduces the complexity of developing hybrid applications.Finally,the feasibility and superiority of the proposed method and system were verified by two sets of experiments.
Keywords/Search Tags:cloud service portfolio recommendation, cross platform cloud service integration, semantic annotation, association degree, SDK automatic generation, interface package
PDF Full Text Request
Related items