Browse Prior Art Database

Publication Date: 2012-May-03
Document File: 7 page(s) / 178K

Publishing Venue

The Prior Art Database


The disclosed article gives out a solution to allow application developers to manage system resources in the multi-tenancy cloud environment efficiently. The solution gives out a framework managing the resources with separated heaps and creating objects according to configuration. The solution includes a method to configure the existing application code with minimum configuration, which is very useful for developers with legacy applications which do not support multi-tenancy before and presents the method to support the configuration using Java annotation or any configuration file format.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 25% of the total text.

Page 01 of 7



The disclosed method and solution relate to systems and methods to support multi-tenancy in cloud environment with as low as possible spatial price and temporal price. Cloud computing allows different application vendors to deploy different applications to the service the cloud computing platform. Along with cloud computing, multi-tenancy is defined as the concept to service multiple isolated applications with a single running instance of application or application container in the cloud platform. Besides reducing spatial and temporal price to service the same number of tenants, multi-tenancy also simplifies administration and provisioning of the tenants. But at the same time, the cloud computing and multi-tenancy also bring ups the challenge of isolation. The isolation is necessary for both security and performance concerns. Take an example of implementing cloud services using Java Application Server which leverages JavaEE technologies; the private application data specific for each tenant should be isolated from other tenants' access, which is a security isolation; the garbage collection of each tenant should be isolated from other tenants' processing, which is a performance isolation. Therefore, the more is shared, the lower the spatial prices and temporal price, but at the same time the more risk for security and performance impact. Then the big problem for the cloud computing and multi-tenancy system design is: how to share the most common data among tenants while isolating tenant-specific data among tenants.

Summary of the disclosed solution
To enable the existing application servers and applications to support multi-tenancy with high efficiency and simplicity, the disclosed solution gives out a method to configure the code of the existing application server or application; and by parsing the configuration data at runtime, the virtual machine is able to create objects in different address spaces for different tenants; and the garbage collection of different tenants are executed independently to release the tenancy-specific resources. The disclosed solution includes a configuration parameter of tenant-intention whose value can be any one of the three values: SHARE, SPECIFIC and INHERIT. SHARE means the object should be shared among tenants and allocated in the System Heap; SPECIFIC means a new object should be created for each tenant and allocated in the tenant-specific Heap; INHERIT means the object address space should be the same as its parent object. By default all tenant-intention values of all objects are INHERIT and if an object has neither tenant-intention argument specified nor its parent specified, its tenant-intention value is set as SHARE. Following the disclosed solution, application server and application developers can enable the existing code to support multi-tenancy with minimum configuration. The following sec...