I see that OpenCL 2.2 has been released and reading through the press release there are a couple of notes that might be of wider interest.
By finalizing OpenCL 2.2, Khronos has delivered on its promise to make C++ a first-class kernel language in the OpenCL standard,” said Neil Trevett, OpenCL chair and Khronos president. “The OpenCL working group is now free to continue its work with SYCL, to converge the power of single source parallel C++ programming with standard ISO C++, and to explore new markets and opportunities for OpenCL — such as embedded vision and inferencing. We are also working to converge with, and leverage, the Khronos Vulkan API — merging advanced graphics and compute into a single API.
Vulkan is a new generation graphics and compute API that provides high-efficiency, cross-platform access to modern GPUs used in a wide variety of devices from computers and consoles to mobile phones and embedded platforms.
There is page of GPU accelerated applications in science applications here
I just came across an interesting paper on cross-platform OpenCL programming. The Hitchhiker’s Guide to Cross-Platform OpenCL Application Development. In particular it highlights a number of issues and offers workarounds. These include Framework bugs, Specification limitations and Program bugs.
There are an increasing number of scientific applications taking advantage of GPU acceleration.
Now in its fourth year, the International Workshop on OpenCL will be held during April at the C3 Convention Center in Vienna, Austria. The majority of sessions will run April 20–21 and will consist of a mix of keynotes, academic papers, technical presentations, tutorials and poster sessions. The workshop kicks-off on April 19 with an Advanced Hands On OpenCL tutorial.
There is a page of scientific applications taking advantage of GPU here.
The 3rd IWOCL (International Workshop on OpenCL) takes place at Stanford University, California from Monday 11 to Wednesday 13 May 2015, and includes the addition of an Advanced Hands-On OpenCL course to the schedule on Monday.
Acceleware will be offering two 4-day training courses in Canada. The first course will be in Calgary Alberta from May 26-29, 2015. The second course will be offered in Montreal, June 9-12, 2015. These professional four day courses are designed for programmers who are looking to develop comprehensive skills in writing and optimizing applications that fully leverage data parallel processing capabilities of GPUs.
The Khronos Group has released revision 30 of the SPIR-V specification. This revision of SPIR-V includes multiple corrections and synchronizes all token spellings to the official headers. These official C/C++ headers are available along with the specification in the registry.
A recently publication “High Performance in silico Virtual Drug Screening on Many-Core Processors” DOI describes porting BUDE (Bristol University Docking Engine) to OpenCL.
Our highly optimized OpenCL implementation of BUDE sustains 1.43 TFLOP/s on a single NVIDIA GTX 680 GPU, or 46% of peak performance. BUDE also exploits OpenCL to deliver effective performance portability across a broad spectrum of different computer architectures from different vendors, includ- ing GPUs from NVIDIA and AMD, Intel’s Xeon Phi and multi-core CPUs with SIMD instruction sets.
BUDE is now one the fastest HPC applications ever developed and nicely demonstrates the portability of OpenCL across different architectures.
There is a list of GPU accelerated applications here.
I know that Fortran is still very important in scientific computing so this may be of interest.
CLFORTRAN is an open source (LGPL) Fortran module, designed to provide direct access to GPU, CPU and accelerator based computing resources available by the OpenCL standard.
Added to the GPU Science page.