Assigning data and performance models to a virtualized service

Relevant for: GUI tests and API tests

When you create and design a virtualized service, part of the configuration process is defining the expected behavior for the service: how quickly it should respond, what the requests and responses to this service should be, and so forth. Because of this, you define performance models and data models for each virtualized service using Service Virtualization. These models are then saved with your virtualization project.

Later, when you add a virtualization project in a UFT One test, the performance models and data models are included with each virtualized service. When you run a test using the virtualized service, you select one of the models to use for a test run:

Performance Models

When you create a virtualization project and add services to the project, you can specify precisely how these services should perform when deployed and run. You can define how quickly the service responds, how often to send requests and responses to the service, the data load for the simulated server, and so forth.

After you create the necessary models in Service Virtualization and add the project to a test in UFT One, you can choose which model to use for each test run.

There are a number of different types of performance models for a virtualization project:

  • User-defined: This model reflects the customized performance settings created in Service Virtualization for a service. Each of the user-defined performance models is available for use in your UFT One test.

  • Offline: This model simulates the unavailability of a service. This model is available for all UFT One tests.

  • None: This model makes the service respond as quickly as possible. This model is available for all UFT One tests.

For details on setting and defining performance models for your service, see the Service Virtualization User Guide

Data Models

In addition to defining performance models for your virtualized service, you can also specify data models. Like performance models, the settings for each model are defined in Service Virtualization, and available for use in UFT One tests.

Data models enable you to customize the requests and responses of the service to simulate real service performance. When you create a virtual service, you define the data model, either by providing the requests and responses for the service, or providing a data source that supplies the request and response values. In addition, you can set rules defining the data source use for each of the services and each of the different models.

All data models are user-defined.

For details on setting and defining data models for your service, see the Service Virtualization User Guide.