Unit test patterns for multicore software

多核软件的单元测试模式

• 批准号: 485144-2015
• 负责人: Labiche, Yvan
• 金额: $ 0.8万
• 依托单位: Carleton University
• 依托单位国家: 加拿大
• 项目类别: Collaborative Research and Development Grants
• 财政年份: 2018
• 资助国家: 加拿大
• 起止时间: 2018-01-01 至 2019-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=656259
• 关键词: Unit; test; patterns; multicore; software

Many industry, including Ericsson, experience increased cost of the software verification activity and one of the reasons is the ever growing size of the testing activity itself, as manifested by the increased number of test (case) executions. One reason for this ever increasing number of test executions is the adoption of agile, cross-functional teams, which, under time pressure, focus on showing that the produced application code works as intended, without necessarily verifying all aspects of the software behaviour. One symptom of this phenomenon is duplication in the test code they write. **This project, titled "Unit test patterns for multicore software", aims at better understanding such duplicated test fragments, often referred to as clones. Detecting code clones in test code and understanding their purposes from a testing point of view will allow us to refactor test code to make the testing activity more efficient and therefore free resources (time of cross-functional team members, test case execution time) to make the testing activity more effective. This should result in better software quality. This will constitute a major contribution to the industry partner of this project, namely Ericsson. From a scientific point of view, this project will contribute to knowledge on test code clones, test code clone detection, and test code patterns. **There is anecdotal evidence that these problems are not specific to the industry partner of this project but are shared by other private organizations in Canada, for instance in the aerospace domain. Public knowledge resulting from this project will therefore be of high interest to a broad sector of the industry in Canada.**The Principal Investigator being in charge of university courses, at the undergraduate level as well as at the graduate level, on software verification, validation and testing, public knowledge resulting from this project will help heighten awareness about the issues addressed in this project, thereby better preparing undergraduate and graduate students with skills that will benefit their future employers in Canada. This should result in higher software product quality for these Canadian employers.

包括爱立信在内的许多行业都经历了软件验证活动成本的增加，其中一个原因是测试活动本身的规模不断增长，这表现在测试（用例）执行数量的增加。测试执行数量不断增加的一个原因是采用了敏捷的跨职能团队，在时间压力下，这些团队专注于显示所生成的应用程序代码按预期工作，而不必验证软件行为的所有方面。这种现象的一个症状是他们编写的测试代码中的重复。** 这个项目名为“多核软件的单元测试模式”，旨在更好地理解这种重复的测试片段，通常被称为克隆。检测测试代码中的代码克隆并从测试的角度理解它们的目的，将使我们能够重构测试代码，使测试活动更有效，从而释放资源（跨职能团队成员的时间，测试用例执行时间），使测试活动更有效。这将带来更好的软件质量。这将对该项目的行业合作伙伴爱立信做出重大贡献。从科学的角度来看，这个项目将有助于测试代码克隆，测试代码克隆检测和测试代码模式的知识。** 有传闻证据表明，这些问题并不是该项目的工业伙伴所特有的，而是加拿大其他私营组织所共有的，例如在航空航天领域。因此，该项目产生的公共知识将引起加拿大工业广泛部门的高度兴趣。主要研究者负责大学课程，在本科阶段以及在研究生阶段，对软件验证，确认和测试，公共知识，从这个项目产生的将有助于提高对这个项目中解决的问题的认识，从而更好地准备本科生和研究生的技能，这将有利于他们未来的雇主在加拿大。这将为这些加拿大雇主带来更高的软件产品质量。