2024 State of the Java Ecosystem

First, let’s look at the most-used Java versions in production.

At one time, Oracle only released new versions with major updates and changes in the JDK. Now, Oracle releases new Java versions every six months—typically in March and September—with each release containing a handful of new features and bug fixes. Every two years, Oracle introduces a new Java long-term-support (LTS) version with updates to help improve stability, security, and performance, which developers often cite as one of the most important factors for upgrading Java versions.

Oracle released Java 21 in September 2023, which marked a significant milestone for Java with notable improvements to preview features like virtual threads and upgraded libraries, as well as advancements to syntax that put Java on par with many more modern languages.

In the six months after the release of Java 21, 1.4% of applications monitored by New Relic were using it. To put this into perspective, in the six months after Java 17 was introduced, only 0.37% of applications were using it, which is 287% fewer.

In addition, the adoption rate of Java 17 far exceeded what the developer world saw when Java 11 was introduced. About a tenth (9%) of applications were using Java 17 in production in 2023, and now 35% of applications are using Java 17, representing a nearly 300% growth rate in one year. It took years for Java 11 to reach anywhere near that level.

Less than 2% of applications were using Java non-LTS versions, which makes sense since they’re usually not used in production.

Now let’s look at the most popular JDK vendors.

Java used to be closed-source or proprietary, which meant developers could only download the JDK directly from Sun Microsystems. However, roughly 10 years after Sun Microsystems released the first version of Java, Oracle released the first open-source Java version. OpenJDK maintains it and allows developer communities and vendors like Microsoft and Amazon to maintain other versions of the JDK.

In 2020, Oracle was the most popular JDK vendor, comprising roughly 75% of the Java market. There was a noticeable movement away from Oracle binaries after the more restrictive licensing of its JDK 11 distribution (before the return to a more open stance with Java 17), and we’ve seen a steady decline year-over-year (YoY) ever since then. While Oracle retained the top spot in 2022 (34%), it slipped to 29% in 2023, and it’s now at 21%—which represents a 28% decrease in one year.

The use of Amazon increased to 31% of the market in 2023 (up from 2.2% in 2020 and 22% in 2022), but has dropped to 18% in 2024, which represents a 43% decrease YoY.

The rising star this year is Eclipse Adoptium, adoption of which rose 50% YoY from 12% to 18%. Because Eclipse Adoptium is community-managed, this JDK tends to be updated more frequently than the Oracle and Amazon JDKs.

Java garbage collectors (GCs) are memory management components used to prevent memory leaks, optimize memory usage, and ensure the overall performance and stability of Java applications.

Some Java versions only support specific GCs, so which GC developers use is partially dependent on their Java version.

Since Java 11, the Garbage-First (G1) GC has been the default. Being the default collector might explain why 43% of customers are using it and why there was a big jump in usage for Java 11, 17, and 21 compared to Java 7 and 8. In addition, one of G1’s primary benefits is that it clears smaller regions instead of clearing large regions all at once, which optimizes the collection process. It also rarely freezes execution and can collect both the young and old generations concurrently, making it a great default for developers.

The second most popular GC is Serial (37%), which is ideal for applications or systems that run on a single processor or where a high number of Java virtual machines (JVMs) are running on the same machine. It also has lower CPU and memory overhead compared to more complex GCs, making it suitable for resource-constrained environments.

The New Relic data shows a 18% YoY increase in applications running with four or fewer cores.

The drive to run smaller makes a lot of sense in cloud environments where people are often deploying containers. But this trend can pose unexpected issues for some applications. In particular, many of the concurrent benefits from the default G1 GC on recent JVMs vanish when running with fewer than two cores. All those single-core instances may as well be using the Serial collector—and paying the performance cost of that.

Any software application or system can have bugs and issues in the testing or production environments, and developers use logging tools to troubleshoot issues and fixes. However, logging is only useful when it provides the required information from the log messages without adversely impacting the application’s performance. Software developers use various logging frameworks to solve these issues. In fact, 91% of the Java applications reporting to New Relic use logging frameworks.

The most-used logging framework was Log4j with 76% of Java applications using it, followed by JBoss Logging (61%) and Logback (52%).

Also, most (83%) Java developers rely on SLF4j, which is a framework that acts as an abstraction for other types of Java logging frameworks. SLF4j enables software developers to use the logging framework of their choice and enables applications to switch to any Java logging framework interchangeably without impacting its implementations or doing any changes. Due to this functionality, SLF4j makes applications independent of logging frameworks, providing more flexibility and portability for logging across any part of the system. It also means that Java applications can use more than one logging framework.

More than two-thirds (41%) of the Java applications reporting to New Relic use encryption libraries: 17% use Bouncy Castle, 16% use Spring Security, and 6% use Jasypt. Bouncy Castle has been a popular library for encryption for many years due to its set of cipher suites and utilities.

While only 0.09% developers use the Amazon Corretto Crypto Provider (ACCP) library, we expect more applications to use it in the near future as companies and developers look to consolidate vendors and because it typically provides better performance.

When looking at databases, Oracle Database is the most widely used, with 17% of Java applications reporting to New Relic using it. Oracle Database is known for its scalability and ability to manage large amounts of data quickly and efficiently. As such, it tends to be the preferred database system for enterprises. Additionally, it offers customer support and a robust set of tools.

The second most popular database system is PostgreSQL, with 14% of Java applications reporting to New Relic using it. While Oracle Database is managed by Oracle directly and available through a license, PostgreSQL is an open-source database that’s free to use and preferred for managing read-write operations and complex queries.

MySQL comes in third, with 13% of Java applications using it. MySQL is also an open-source database. It offers fewer features than Oracle Database and PostgreSQL, which makes it more stable and faster at processing, especially when processing read-only queries.

PostgresSQL has gained popularity in the last year because it’s more compliant with the SQL specs and it supports more features than MySQL. MySQL is still a top database system, but we can see that its popularity is decreasing while PostgresSQL’s popularity is growing.

We also looked at what kind of Java-related questions and requests developers are asking via the New Relic AI GenAI observability assistant. Since January 2024, the data shows that 34% of the 483 Java-related questions were how-to type (learning) questions, 21% were related to querying for a specific metric, 14% were about configuration, and 6% were about troubleshooting.

Sample questions and requests include:

• How do I log API calls in my Java Spring Boot?
• If the Java framework changed from Tomcat to JBoss, do I need to reconfigure the agent?
• Create a dashboard for container memory utilization vs. JVM heap utilization.
• This monitored application uses the Vaadin framework in Java. It has high memory consumption. What could it be?
• In JVM, what does G1 Eden Space heap usage mean?