Project Valhalla Confirmed for JDK 28 After Decade of Development
Original: Project Valhalla, Explained: How a Decade of Work Arrives in JDK 28
Why This Matters
Valhalla could fundamentally improve Java performance for memory-intensive applications and large-scale data processing workloads.
Oracle engineer Lois Foltan confirmed on June 15 that JEP 401 (Value Classes and Objects) will integrate into OpenJDK, targeting JDK 28. The 197,000-line code addition across 1,816 files launches as a disabled-by-default preview feature, marking the first phase of the long-awaited project.
Project Valhalla aims to enable Java classes to be written normally but executed with primitive-like efficiency, addressing a core Java limitation. The project emerged from 2014 observations that Java objects, stored as heap references, create memory inefficiency at scale. Every object carries metadata overhead and requires garbage collection, while arrays of objects become scattered pointers rather than dense data. Modern hardware exacerbates this: CPUs are two orders of magnitude faster than main memory, making cache efficiency critical. Valhalla addresses this by allowing value classes—objects that can be laid out densely in memory like primitives. Oracle engineer Brian Goetz cautioned that this JDK 28 integration represents only the first phase. The pull request alone spans 1,816 files and adds over 197,000 lines of code, prompting committers to postpone other major changes during integration. The feature ships disabled by default as a preview, not a full release. The project faced industry skepticism over its decade-long timeline, with community jokes about reaching Norse Valhalla before the Java feature shipped. Goetz noted critics would transition from "they'll never ship it" to "they didn't ship the important part," acknowledging the phased rollout approach.