Artificial Intelligence (AI) and Machine Learning (ML) technologies have evolved over the years and their use in automated testing in now beneficial in many ways. The adoption of AI in microservices testing has allowed organizations to drive better results and gain greater efficiency. AI has even redefined how microservices-based applications are tested.

人工智能（AI）和机器学学习（ML）技术已经经历了好多年的发展，如今它们也以多种方式在自动化测试的使用中发挥作用。人工智能在微服务测试中的使用已经使得相关的组织获得更好的测试结果以及更快的效率。人工智能甚至已经重新定义了基于微服务应用的测试。

Using AI in software testing help both developers and testers by improving accuracy and gaining time. Automated testing increases the depth of the tests performed, allowing more test coverage.

在软件测试中使用人工智能，通过提高准确性和节省时间能同时帮助开发人员和测试人员。自动化测试增加了测试的深度，能够覆盖更多的测试需求。

AI-based software testing is now used for test creation, test execution, and data analysis; hence, increasing efficiency and improving test accuracy and coverage. It also allows for easier test maintenance, especially for managing test data. AI can be useful for effective data modeling and for root-cause analysis. With automated tests, companies save both time and money.

以人工智能为基础的软件测试现在用于测试新品，测试运行以及数据分析；因此，提高了效率和改善了测试准确性、覆盖率。测试的维护也更加简易，特别是对于测试数据管理而言。人工智能在有效的数据模型和根本原因分析当中是游泳的。由于自动化测试，企业不仅节省了时间还节省了金钱。

Canary testing – minimal tests which quickly and automatically verify that everything you depend on is ready – is extremely useful when it comes to microservices-based applications testing. AI can improve the automation of canary testing in microservices-based applications, by using concepts such as deep learning to identify the changes in the new code and the issues within it or even comparing the experience of a small group of users with existing users. Everything can be done automatically and there is no need for human intervention.

金丝雀测试-用极小的测试快速并且自动化的验证所有依赖的东西已经准备就绪-当用在以微服务为基础的应用测试上非常的有效。人工智能能够改善基于微服务的应用的金丝雀自动化测试，通过使用一些概念，如深度学习鉴别新代码的变化，又比如存在的问题或者是现存用户与一小部分使用者体验的对比存在的问题。所有的事情都能被自动的解决而不需要任何的人为干预。

However, there are a few challenges to AI-based testing. Indeed, functional and unit tests are easy to automate, it is not necessarily the case for integration tests, because of their complexity.

然而，在以人工智能为基础的测试中也存在一些挑战。事实上，功能测试喝单元测试是容易实现自动化的，对于像集成测试这样的案例是没有必要的，因为它们的复杂性。

Moreover, testing microservices-based applications with an AI-drive approach needs considerable technical expertise from testers. Companies should have testers who are able to use AI-based tools required specifically for microservices-based applications. In microservices test automation, testers also need to determine which AI use cases are the best. Depending on the goal, they will have to either use AI for creating unit tests by having AI performing static code analysis and determining the portions of code not covered by the unit tests. Or they can use AI to keep the unit tests up to date whenever the source code changes.

并且，用人工智能驱动的方式测试基于微服务的应用需要考虑测试人员的技术专业度。企业应该有能够使用专门针对基于微服务的应用程序所需的基于AI的工具的测试人员。在微服务自动化测试中，测试人员也需要决定，使用那种人工智能技术是最好的。基于目的，他们将不得不使用AI通过创建AI执行静态代码分析并确定未包含在单元测试中的代码部分来创建单元测试。或者，只要源代码发生更改，他们就可以使用AI使单元测试保持最新。

Therefore, AI-based test automation of microservices can generable more reliable and efficient tests, which make testers gain time in test creation, maintenance, and analysis. Using AI for microservices test automation will not enhance every part of software testing but it will make testing faster, smarter, and more effective in the long run.

因此，基于AI的微服务测试自动化可以生成更可靠，更有效的测试，这使测试人员可以在测试创建，维护和分析中节省时间。将AI用于微服务测试自动化不会增强软件测试的每个部分，但从长远来看，它将使测试更快，更智能，更有效。

龙测科技，自动化测试AI驱动，我们在行动。