Research And Design Of Unit Test Cloud Platform

Code Test System(CTS)is a kind of stand-alone software which can realize automatic unit test.However,the existing single-machine version of CTS is not stable enough to support multi-threading at infrastructure,and has the problem of low efficiency in testing complex and large proj ects,as well as common problems traditional native testing software suffers,such as complex installation,unfriendly user experience,waste of human and material resources,and inability of testing by multi-users in a collaborative way.In order to solve the above problems,this paper mainly completes the following work:1)Based on the research of cloud testing,this paper proposes upgrading CTS from single-machine version to cloud testing service.The cloud version of CTS referred as unit test cloud platform is designed and developed.The framework consists of virtualization layer,cloud platform layer and cloud application layer.This version allows users to obtain code testing service directly via the browser without having to install software on local computer.2)The unit test cloud platform is implemented in this paper.Firstly,in combination with virtualization technology,Hadoop cluster was built to realize the basic support module and resource scheduling module of cloud platform layer.Secondly,using the Spring Boot as the framework and MVC as the design pattern,the web program of the cloud application layer is implemented to provide test services directly to users.Then the business logic processing of the cloud application layer is realized by providing the interface of each functional module of CTS.Finally,the auxiliary test program is implemented to achieve the startup of the testing service with the command line.3)The cloud platform is tested.Based on single-user scenarios and multi-users scenarios,functional testing is performed on each module of the platform to verify its usability and integrity.In addition,through tesing aginst real maj or proj ects,it is proved that the testing result is accurate and testing process is efficient.