Browse Prior Art Database

Software model and methods generating self-contained testcases packages during real time data processing for computational data-centric applications .

IP.com Disclosure Number: IPCOM000226343D
Publication Date: 2013-Mar-28
Document File: 4 page(s) / 58K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is a method to combine into a single step all the debugging steps taken to analyze application issues during the production or pre-production stage. The novel contribution of the proposed design is to extend the application execution mode by providing an on-demand automated generation of self-contained test case packages.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 32% of the total text.

Page 01 of 4

Software model and methods generating self-contained testcases packages during real time data processing for computational data-centric applications .

The current process for analyzing application issues during the production or pre-production stage comprises: analyzing specific use cases and requesting and reviewing logs and trace files; reproducing the failure if possible, debugging, and fixing the issue in the development or quality assurance (QA) environment; and implementing a unit test case for a regression test suite. In many cases, these steps require a lot of engineering, time, and resources. A method is needed to simplify

debugging computational data-centric application functions, minimize the efforts in resolving customer issues, and optimize the software quality process by ensuring adequate test coverage.

Many available tools can auto-generate test cases based on the source code, interaction with the user interface (UI), or processing database (DB) specific transactional data. However, the available tools only generate unit tests or test cases that might not always be efficient for detecting the software defects , especially in computational data-centric applications where dynamic datasets from various data sources are being processed.

The invented solution is to combine all the debugging steps identified above into one, through an automated generation of the test cases , producing self-contained packages at the place where the error or issue has been reported.

The novel contribution of the proposed design is to extend the application execution mode by providing an on-demand automated generation of self-contained test case packages. These packages include generated test case codes and built -in referenced runtime data object snapshots. The test case build is executed at the right place, at the right time (i.e., in the runtime environment, during application execution when all the processed data are available regardless of their data sources). Executed as standalone test cases or applications, the test cases can be used to further analyze and debug the issue, validate the fix, and then be included into a QA regression test suite.

The proposed design is efficient if applied to the following application execution scenarios:

¡⁄ The application is integrated with modules written in different computer languages

¡⁄ The input datasets are specific to a particular customer and cannot be predicted and/or tested during the development phase

¡⁄ The full data set is not available for reproducing the issue because sensitive data is present in the customer database

¡⁄ Input datasets are being generated dynamically based on the data retrieved from external systems (e.g., messaging, UI input, optimization engine, etc.)

¡⁄ The runtime environment cannot be accessed through the developer 's debugging tool

Regardless of whether the issue is reported at the customer site, implementing the invented solution improves software...