Emulation of cloud-scale environments for scalability testing

Cloud computing increases the level of connectivity between software applications. IT management applications delivered as a service may need to connect to tens of thousands of endpoint systems. In order to validate the application's reliability and performance at these very large scales, its scalability needs to be tested before being deployed in the cloud. We use an emulation approach, whereby endpoints are modelled and then executed in an emulation environment, which we call 'Kaluta'. The key aspect is to balance the modelling of the endpoint systems such that it is rich enough to 'fool' an unmodified application-under-test into thinking that it is talking to real systems, but lightweight enough such that tens of thousands of instances of model systems can be executed simultaneously in the emulation engine. We present an industry case study---CA IdentityMinder(TM)-as-a-Service---to demonstrate the effectiveness of using emulation to validate the scalability of a cloud hosted application.