Technology NEWS
Users of the Bazel build tool can now enjoy the benefits of Build Scans. Gradle Enterprise leverages Bazel’s native build event service and remote cache functionality, which allows Bazel builds to publish a Build Scan to Gradle Enterprise without any significant build modifications. Bazel Build Scans provide shareable access to the console logging, build performance profile, action cache hit rate, requested target results and failure diagnostics, effective build configuration options, build environment and detailed individual test results. The Tests Dashboard can be used to identify the slowest, most unstable, and flakiest individual tests across many builds based on different search criteria. Failures, Performance and Trends dashboards are supported as well.
Predictive Test Selection intelligently skips irrelevant tests to speed up developer feedback and reduce CI costs, using machine learning. This system automatically learns a test selection strategy by analyzing historical code changes and test outcomes. Gradle Enterprise now includes a Predictive Test Selection simulator, in beta, which demonstrates the benefits and risks of test selection for a customizable set of test builds, so that you can make a data-driven decision as to which projects and builds should adopt it.
To learn more about these and other new capabilities available in 2021.4 check out the release notes.
Success Story
Gradle Enterprise has emerged as the de facto standard tool for build and test data analytics and as a source of acceleration technology for many of the most important open-source projects, including Spring, JetBrains Kotlin, JUnit, Spock, JHipster, Ratpack, and now Micrometer.
Micrometer's build has been enjoying the advantages of the Gradle Build Tool since the inception of the project. This summer the Gradle Team provisioned a Gradle Enterprise instance which helped the Micrometer team substantially improve the project’s build performance and their troubleshooting process.
In this guest-authored Blog by Jonatan Ivanov, a senior software engineer on the Spring Team @VMWare working on Micrometer, Jonatan discusses how they achieved major developer productivity improvements leveraging key DPE tools available in Gradle Enterprise such as Build Scan™, Build Cache and Performance and Trends dashboards. For example, build times improved locally and on CI from roughly 2 minutes to 5 seconds when using local and remote build caches.
Expert Take
In this recent Developer Productivity Engineering Lowdown (on-demand webcast), we talked to senior software engineer Grant Jenks from the Developer Insights team, which is part of the LinkedIn Developer Productivity and Happiness organization. Developer Insights works like a “Fitbit for engineering teams'' to identify and improve pain points in developer workflows. Grant explores the platform they built to improve the productivity of their engineers with the goal of shipping new and better features to LinkedIn users.
Best Practices
Check out our new tutorial playlist where you’ll learn how to get started with Build Scans and get quick wins using all the major Build Scan functionality. We designed the series in 1-minute bite-sized chunks so you can see and realize the value quickly. Here are the playlist video titles:
- Automating the Build Scan
- Sharing the Build Scan
- View and Share Console Log Output
- Detect and Correct Deprecations
- Build Timeline to View Build Performance
- Performance View to Analyze Feedback Speed
- Test View to Assess Test Performance
- Dependencies View to Understand Related Failures
- Plugins View to Assist in Build Debugging
- Add Custom Values to Build Scans
- View Switches and Infrastructure Data
Upcoming Featured Event
Our next Introduction to Developer Productivity Engineering online workshop will take place on January 14, 9:00 am - 11:30 am PT. In this hands-on online educational event with labs, DPE guru, Raju Gandhi, will demonstrate DPE best practices and tools in action—using examples from Java projects that rely on Maven or the Gradle Build Tool. In this workshop you will learn how to:
- Accelerate debugging and root cause analysis time in half leveraging Build Scans.
- Achieve up to 90% reductions in local and remote build times using Build Cache and Test Distribution.
- Avoid performance regressions, continuously improve toolchain reliability, and better manage flaky tests with build performance and build failure analytics.
|
