| Popis: |
Scientist developers are developing systematic testing techniques to assure software quality, routinely. A key challenge is the oracle problem, a situation in which appropriate mechanisms are unavailable for checking if the code produces the expected output when executed using a set of test cases (TCs). Metamorphic testing (MT) alleviates the oracle problem by specifying the relationship that a source TC and its follow-up TC shall meet. Such relationships are called metamorphic relations (MRs) which are necessary properties of the intended program’s functionality. Through checking whether the MRs hold or not, MT, a property-based software testing technique, can be applied to overcome the oracle problem.Scientific model developers are able to verify and validate their software via MT, even when the expected output of a given TC is not readily available. The tenet is to check whether certain relations hold among the expected outputs of multiple related inputs. Based on MT and MRs, the overall objective of this dissertation is to achieve a function as a service (FaaS) architecture by developing and deploying a new class of services called metamorphic software services. These services, expressing different relations between software outputs, will serve a broad range of stakeholders in a similar way that MRs serve software testers. To that end, the first component of this research is to develop useful metamorphic software services which are based on the research of the MRs. After obtaining useful metamorphic software services for the community of Storm Water Management Model (SWMM, a dynamic rainfall-runoff simulation model that computes runoff quantity and quality from primarily urban areas), the second component is to engage social media information in metamorphic software service discovery. Finally, the third component aims to build and deploy the metamorphic software services in a FaaS way, so as to provide a serverless platform for SWMM communities. This dissertation work will be able to provide a serverless platform for stakeholders to test SWMM and its integration, and improve the efficiency of extracting MRs and applying MT to SWMM. It also provides the cornerstone for future expansion to more software which can extend the capabilities of the water domain-experts (e.g., hydrologists and sewer district engineers) to perform scientific computations, process complex information, and make informed decisions. |
