Key Takeaways from JavaOne 2017: JDK9 Features, Java EE Updates, and Serverless FAAS
The article summarizes the JavaOne 2017 conference, highlighting JDK9's modular system and collection factories, Java EE8 enhancements, JVM improvements, a graph‑based recommendation demo using EE4J, and the rise of serverless FAAS as a cloud‑native computing model.
In October 2017 a group of Ant Financial engineers attended Oracle's JavaOne 2017 conference in San Francisco, where the latest Java platform innovations, cloud‑native services, FAAS, micro‑services, big data, AI and blockchain were presented.
JDK 9 New Features – The most significant change is the introduction of the modular system (Project Jigsaw), which restructures the standard library, removes rt.jar , tools.jar and the traditional jre/ directory, and replaces them with conf/ and jmods/ . The module system also encapsulates internal APIs such as sun.misc.Unsafe . Collection factory methods like List.of() , Set.of() and Map.of() simplify object creation, e.g.: Set<String> set = Set.of("a", "b", "c"); which replaces the verbose pre‑Java 8 code that manually instantiated a HashSet and added elements.
Improved Stream API – New methods such as takeWhile() and dropWhile() (inspired by Scala) allow predicate‑based element skipping. Example: Stream.of(2, 4, 6, 8, 9, 10, 12) .dropWhile(n -> n % 2 == 0) .forEach(System.out::println); prints 9 , 10 , 12 . Another example demonstrates a concise iterator style: IntStream.iterate(1, i -> i < 100, i -> i + 1) .forEach(System.out::println); outputs numbers 1‑99.
JVM Enhancements – G1 becomes the default GC in JDK 9, while Azul’s Zing JVM introduces the C4 (Continuously Concurrent Compacting Collector) algorithm to eliminate long‑stop‑the‑world pauses. The Files.lines() method now uses memory‑mapped I/O for higher performance, fixing a BufferedReader bottleneck (see JDK‑8072773).
Java EE 8 Updates – Oracle highlighted new APIs: JSON Binding, JAX‑RS reactive client, HTTP/2 support in Servlets, additional security APIs, and fresh features in Bean Validation and CDI.
Recommendation Engine Demo – A use‑case relevant to Ant Financial demonstrated building a content‑recommendation engine with EE4J, JNoSQL, and the TinkerPop graph framework, showing how a graph database (e.g., Neo4j, ArangoDB) can simplify DAO‑based graph operations compared with traditional SQL.
Serverless FAAS and Cloud Computing – The conference emphasized serverless Function‑as‑a‑Service as a deeper form of cloud hosting: developers write only business functions while the platform handles provisioning, scaling and routing. Advantages listed include easy adoption, treating Java functions as first‑class citizens, event‑driven statelessness, theoretical unlimited scalability, simplified deployment, and cost benefits. The article also notes the risk of vendor lock‑in due to differing provider APIs.
Additional sessions covered using Docker, Kubernetes and OpenShift for Java application deployment, but details are omitted.
Overall, the talks delivered a broad view of modern Java development, from low‑level JVM improvements to high‑level serverless architectures. Presentation materials can be downloaded from the official conference site .
AntTech
Technology is the core driver of Ant's future creation.
How this landed with the community
Was this worth your time?
0 Comments
Thoughtful readers leave field notes, pushback, and hard-won operational detail here.