定制
为了提供完整的功能,本网站需要您明确同意存储浏览器Cookie。如果您不允许使用cookie,您可能无法使用网站的某些功能,包括但不限于:登录,购买产品,查看个性化内容,切换网站语言。建议您允许所有的Cookie。
未绑定有效的TSI角色,无法登录
搜索
By Jake Wardon, TSI Software Engineer III
As of this writing, TSI has begun to achieve DevOps and Continuous Delivery on some of our newer products. We’ve done this by setting up builds and deployments that are automatically triggered by events in source control.
For instance, when a pull request is created, we run automated checks (code quality analysis, unit tests, and end-to-end tests) and report the results back to source control, so reviewers can see whether the tests have passed without having to pull down that branch, compile, and test—making PR reviewing much quicker than it was previously. When a PR is merged, a build is automatically triggered—and in some cases, delivered—to the end user within minutes.
As we close out the year, we look forward to enjoying the efficiency gains realized by our hard work in the DevOps space, as well as incorporating automation into more of our products’ lifecycles.
One of our greatest challenges will be integrating these concepts in our diverse ecosystem of product types. For instance, the tools and methods we use to delivery a SaaS product, will not be the same as those we’d use to delivery desktop or mobile software—and those will be different, still from how we deliver firmware updates to devices in the field.
Read Part 1 of Jake's series. In future articles, Jake will explore the various elements of DevOps in more depth—including the tools and processes that make DevOps possible, and concepts such as “Left Shift” that we strive to achieve.