Automatic Query Rewriting Schemes for Multi-Tenant SaaS Applications

Abstract

In Software as a Service (SaaS) environments, designing and realizing multitenant schema-mapping that supports a shared database with custom extensions is a non-trivial task. Universal Table is one promising schema-mapping technique that is commonly used. However, there has been little research devoted to the design and realization of a query rewriting scheme for Universal Table. In this paper, we present a collection of general query rewriting schemes for Universal Table that can transparently transform tenant-specific logical queries into corresponding physical queries. Based on the design, we have developed a Java-based schema-mapping and query rewriting middleware for Universal Table and a sample online shopping SaaS application to verify its feasibility. Additionally, analytical results that can be used to predict the overhead of our schemes are also reported. Finally, we conduct a series of experiments and find that the results not only agree well with our analytical predictions but also show that our schemes are scalable to the number of tenants and the number of concurrent database connections.