New Parallel Java Initiative
The Heterogeneous System Architecture Foundation (HSA) has announced an initiative to bring better parallel processing capabilities to Java so that Java programs will be able to access the potential potential of GPUs and other recent HPC innovations. HSA is a consortium of companies (including AMD, ARM, and others) committed to "building a heterogeneous compute software ecosystem which is rooted on open royalty free industry standards."
HSA president Phil Rogers rolled out the new concept at the Hot Chips symposium in Palo Alto, CA. According to a report at PC World, the new Java specifications are slated to appear with Java 9 in 2015.
Making Java more parallel will help to employ the many legions of Java programmers as chips become more powerful and parallel computing becomes more common throughout the industry. The efforts of SoC vendors like AMD and ARM to encourage standards-based, uniform interfaces will serve to reduce the importance of vendor-controlled software silos and put them emphasis back on the hardware.
Related content
-
Discovering ROCm
AMD’s ROCm platform brings new freedom and portability to the GPU space.
-
Floodlight: Welcome to the World of Software-Defined Networking
Software-Defined Networking (SDN) marks a paradigm shift toward a more holistic approach for managing networking hardware. The Floodlight OpenFlow controller offers an easy and inexpensive way to experience the power of SDN.
-
OpenFlow and the Floodlight OpenFlow Controller
Software Defined Networking (SDN) marks a paradigm shift toward a more holistic approach for managing networking hardware. The Floodlight OpenFlow controller offers an easy and inexpensive way to experience the power of SDN. -
Enterprise job scheduling with schedulix
Cron is the simplest job scheduler on Unix and Linux systems, but if you are looking for an enterprise-level solution that offers unique, sophisticated sequencing and monitoring functions, schedulix might just fit the bill. -
Hadoop for Small-to-Medium-Sized Businesses
Hadoop 2.x and its associated tools promise to deliver big data solutions not just to the IT-heavy big players, but to anyone with unstructured data and the need for multidimensional data analysis.