The versatile Java programming environment is getting a speed boost.
An IBM executive said today that graphics processing unit (GPU) acceleration is coming to the Java. That could improve processing performance as much as tenfold for Java-related computing.
IBM’s John Duimovich, speaking in a keynote address at Oracle’s JavaOne conference today, said that GPU accelerators pack considerable nongraphics computing power because of their parallel design, with many subprocessors operating at the same time.
Sumit Gupta, the general manager of accelerated computing at Nvidia, wrote in a blog post that Java and GPUs will open up a world of opportunities for faster web performance.
“Millions of developers rely on the Java programming language for Web 2.0, big data analytics and scientific computing,” Gupta wrote. “It’s widely used in large-scale distributed frameworks, like Apache Hadoop, due to its ease of programmability, modularity and multiplatform support.”
Speedups vary greatly when Java programmers take advantage of existing GPU compute libraries, based on Nvidia’s CUDA programming environment. Those speedups range from two times to 48 times faster.
According to Duimovich, the company will enable IBM runtimes for server-based GPU accelerators and explore acceleration in ordinary workloads under existing applications programming interfaces.
The upshot, Gupta wrote, is that “this will allow millions of Java developers to accelerate a broad range of applications using GPU accelerators – and achieve speedups that will dramatically improve the capabilities of the applications. Plus, the acceleration will fuel a new generation of Java-based enterprise applications that would not have been possible without GPUs.”
Such programs will result in everything from high-performance distributed fraud detection and financial analysis, to high-throughput video and image analytics and modern scientific applications.