After six months of development, Oracle released the Java SE 16 (Java Platform, Standard Edition 16), which uses the OpenJDK open source project as a reference implementation. Java SE 16 retains backward compatibility with previous releases of the Java platform, all previously written Java projects will work without changes when launched under the new version. Ready to install Java SE 16 assemblies (JDK, JRE and Server JRE) prepared for Linux (x86_64, AArch64), Windows and macOS. The OpenJDK project reference implementation Java 16 is completely open source under the GPLv2 license with GNU ClassPath exceptions allowing dynamic linking with commercial products.
Java SE 16 is categorized as a regular support release category, with updates for which will be released until the next release. Java SE 11 should be used as the Long Term Support (LTS) branch, and updates will be released until 2026. The next LTS release is slated for September 2021. Recall that starting with the release of Java 10, the project moved to a new development process, implying a shorter cycle of forming new releases. The new functionality is now being developed in one constantly updated master branch, which includes ready-made changes and from which branches are branched every six months to stabilize new releases.
In preparation for the new release, development has been moved from Mercurial source control to Git and a collaborative development platform GitHub . The migration is expected to improve the performance of repository operations, increase storage efficiency, provide access to changes throughout the project history, improve code review support, and leverage APIs to automate workflows. In addition, the use of Git and GitHub makes the project more attractive for newbies and developers who are accustomed to Git.
From innovations Java 16 you can mark :
- Added an experimental jdk.incubator.vector module with an implementation of the Vector API, which provides functions for vector computations that are performed using vector instructions of processors x86_64 and AArch64 and allow simultaneous application of operations to several values at once (SIMD). Unlike the scalar auto-vectorization capabilities provided in the HotSpot JIT compiler, the new API allows explicit vectorization control for parallel data processing.
- JDK and VM HotSpot code written in C ++ is allowed