Font Size: a A A

Research And Application Of Data Management And Store For Multi-tenants SaaS Applications

Posted on:2015-12-03Degree:MasterType:Thesis
Country:ChinaCandidate:H WangFull Text:PDF
GTID:2298330422982063Subject:Computer application technology
Abstract/Summary:PDF Full Text Request
With the promotion of cloud computing, SaaS as a new software delivery mode becomemore and more popular. In SaaS way, customers get software service from Internet, instead ofinstalling and deploying it on local server. Multi-tenants is one of the most importanttechnologies of SaaS. Enterprises who rent service from software provider are tenants. SaaSdelivery mode decides cloud computing not provides service to only one tenant. Many tenantsshare a software instance, multiplexing of infrastructure maximize resource utilization, reduceoperating costs.Reasonable and efficient data storage and management is the foundation of the tenantSaaS applications. In this paper, proposed a hierarchical model of data management formulti-tenant SaaS applications and an improved schema mapping technology to solve theseproblems:1. How to meet tenants’ requirements as their business data structures is different.2.Ensure data retrieval performance, even sum of tenants and volume of tenants’ data grows.3.Store tenants’ data on shared devices physically, but keep them isolate logically. Specificresearch work and achievements are as follow:First, designed a hierarchical model of data management for multi-tenant SaaSapplications, which added a schema mapping layer between tenant’s logical view and physicaldata storage layer to manager tenants meta data, switch data query language, reconstructquery results and transfer tenants data. SaaS application request tenants data from uniformdata interface and don’t care how the data store and map.Second, proposed an improved data mapping technology—General Extension TableMapping, which was based on Chunk Folding Mapping. Compare performance of GeneralExtension Table Mapping with other exist mapping technology on storage utilization, retrievalspeed and reconstruct complexity.Last, proved that the hierarchical model of multi-tenants’ data management is feasibleand improved mapping technology is better than other mapping mechanism on performanceby using it on campus energy cloud.
Keywords/Search Tags:SaaS, multi-tenants, data storage for multi-tenants, schema-mapping
PDF Full Text Request
Related items