Font Size: a A A

Research On Multi-tenant Data Management Model Based On XML Schema

Posted on:2014-12-04Degree:MasterType:Thesis
Country:ChinaCandidate:C C YuFull Text:PDF
GTID:2268330401462182Subject:Computer applications
Abstract/Summary:PDF Full Text Request
Cloud Computing is provided with better extensibility, lower cost and flexibilitycapacity, which make SaaS as a new software mode and is paid more and moreattention, especially the small and medium-sized enterprises. SaaS gradually becomean important way for the application of advanced technology.Multi-tenant technology is a kind of only one application instance, leased tomultiple tenants, sharing of software development framework between all tenants. Atmulti-tenant data management, tenants by custom or custom functions, achieve thepurpose of personalized business data extension. Service providers share theresources from the operating system to multiple levels of data structures bysupporting multi-tenant shared storage architecture, which take full advantage of thehardware, database and other resources that greatly reduces the operating costs.But, this kind of mode that tens of thousands of tenants share an applicationinstance also bring a very big problem to the tenants at data management. Because ofthe different business areas of the tenants’ application and business data exists greaterspecificity, the platform of SaaS require to support multi-tenant on the basis ofcommon data needs and meet unlimited extension requirement of the personalizeddata of the multi-tenant. Although the model of data extension based on XML schemahas unlimited extension, it is first to extraction of the XML when inquire data, whichcause the query performance degradation. In order to be able to achieve a goodscalability and good query performance management model of multi-tenant data. Thisarticle does the following work:First, in order to satisfy the multi-tenant data management and extension needsin cloud computing environment, In-depth analysis on the basis of the extendedmodel of XML data, combined of metadata management thinking, the model of dataextension based on XML schema is proposed in cloud computing environment. Themodel contains data custom functions, index customization functions, metadataservices and log analysis services, which can meet the needs of management andextension of multi-tenant data.Secondly, in order to further improve the efficiency of the data query of the multi-tenant data management model, methods put up to enhance the efficiency ofdata query and algorithm. A Personalized index method of tenants customized wasproposed to take advantage of the index of the database for better query efficiency. Arecommendable index for the algorithm was proposed to analyze which are most inneed of establishing the index data items according to tenants’ searched record andwas recommended tenants establish index which can improve the efficiency of query.Finally, in order to verify the feasibility of the proposed multi-tenant datamanagement model, a simulation environment of CloudSim was established.Experimental analysis from the efficiency of query and update data and the resultsshow that the model can not only ensure the extensibility of the data, but also with ahigher efficiency query.
Keywords/Search Tags:SaaS, multi-tenant, multi-tenant data storage, multi-tenant index, indexrecommended
PDF Full Text Request
Related items