Successes and Challenges in Bringing Performance to Java with Inline Types
Let’s consider the following:
• Hardware has evolved since the 90s when Java was first released and performance in Cloud environments is of utmost importance nowadays. • Java can be strengthened by new language features that enhance the developer experience, without sacrificing performance.
These statements underlie the continuous efforts behind Valhalla, a project that seeks to advance Java performance by improving memory layout efficiency and also enhances abstraction, encapsulation, safety, expressiveness, and maintainability in the Java language. The core feature behind Valhalla that drives these benefits is Inline Types, also known as Value Types. In a nutshell, Inline Types can be thought of as special classes that are faster/cheaper objects or programmable/richer primitives.
In this talk, we’ll look at some successes in how Inline Types have been designed and implemented to align with modern hardware and extend the Java language. We’ll also touch upon the challenges of introducing such a fundamental change to a long-standing programming language like Java.
|Full talk proposal (MoreVMs21_talk_3.pdf)||364KiB|
Hey! I’m a Software Developer working on the OpenJ9 VM team at IBM. I’m a big fan of open source and building community, and try to regularly find opportunities to get involved. When I’m not getting lost in my browser tabs, I enjoy spending time outdoors, bouldering and doing aerial silks.
Tue 23 Mar Times are displayed in time zone: Greenwich Mean Time : Belfast change
15:00 - 16:30
|The Strange and Wondrous Life of Functions in Ř|
Jan JečmenFIT CTU Prague, Olivier FlückigerNortheastern University, Sebastián KrynskiCzech Technical University, National University of Quilmes, Jan VitekNortheastern University / Czech Technical UniversityMedia Attached
|Successes and Challenges in Bringing Performance to Java with Inline Types|
Sharon WangIBMMedia Attached File Attached
|Towards a Synthetic Benchmark to Assess VM Startup, Warmup, and Cold-Code Performance|
Stefan MarrUniversity of KentMedia Attached