Last year, Intel acquired FPGA-focused Khronos member Altera. Intel has now announced a new line of hybrid chips that combine FPGAs with their well-known CPUs. One of the more interesting aspects of the new Intel FPGA ecosystem is the Acceleration Stack, an OpenCL based programming environment that can be used by developers for hybrid cards or discrete cards, including FPGAs, CPUs, and GPUs. The stack abstracts the programming required for the FPGAs to streamline and speed up development for accelerators and applications being used. Additionally, it allows for code to be reusable — porting between FPGAs/GPU/CPU should be possible without major changes. OpenCL, a C based programming language, will. This is quite the opposite of what had been available when Intel released the E600C seven years ago.
Hybrid CPU-FPGA devices are expected to see widespread adoption. Intel is concentrating on the programming environment so the same tools will be used whether the CPUs and FPGAs are discrete or hybrid in the same socket. This is called the Acceleration Stack for Intel, and it is a complete programming environment that is based on OpenCL, the common higher level programming language that is converged to Verilog and VHDL for FPGAs. Learn more about the roadmap Intel has working on.
Codeplay has announced that ComputeCpp Community Edition is now available on Windows. It is now possible to develop SYCL applications using Windows and Visual Studio. The Windows release of ComputeCpp CE currently supports Windows 7 or 10 and can be used with Visual Studio 2015. Similar to our Linux version, the hardware you want to use with ComputeCpp you is required to have SPIR OpenCL drivers in order to be supported.
Announcing that the 6th International Workshop on OpenCL will take place on the 14-16 May, 2018 at St Catherine's College, Oxford, UK and that the Call for Submissions is now open. Submissions related to any aspect of using OpenCL (including SYCL, Vulkan Compute and OpenCL based libraries) are of interest, including (but not limited to): case-studies of their use in applications, software tools, programming methods, debugging, performance analysis, and integration.
The Xilinx software defined development environment, SDAccel, is now available on Amazon Web Services for use with Amazon Elastic Compute Cloud F1 instances. SDAccel automates the acceleration of software application written in C, C++ or OpenCL by building application-specific FPGA kernels for Amazon EC2 F1.
Non-profit organization The Blender Foundation has released Blender 2.79, an update to its cross-platform, open-source 3D graphics tool. The new build further improves its Cycles Rendering feature, bringing feature parity with NVIDIA CUDA and improved performance to AMD OpenCL hardware.
A pan-European project has started this month to bring together the technologies needed for exascale computing, tackling the key challenge of power usage. The project started this month, bringing together three existing exascale projects on FPGA accelerators, interconnect and 3D chip technologies to reach performance of 10^18 operations, 10 times that of today's fasest supercomputers. At the University of Manchester they are working on OpenCL as the programming model to configure modules that can be plugged into a system as an HPC accelerator.
Renesas Electronics announced their collaboration to deliver ComputeAorta™, Codeplay’s OpenCL open standard-based software framework for Renesas R-Car system-on-chips (SoCs). The new framework is designed to support software development for the R-Car’s latest image recognition IP, the IMP-X5, a multi-threading core optimized for computer vision and cognitive processing. Codeplay will also provide R-Car with ComputeCpp™, an implementation of the SYCL™ open standard, enabling single source C++ software for high level and object-oriented programming. The result of this collaboration provides developers with standard software development tools and support for a wide range of open source computer vision or open source deep learning software, such as TensorFlow™ library.
NVIDIA Nsight Visual Studio Edition for Microsoft Visual Studio allows you to build, debug, profile and trace heterogeneous compute, graphics, virtual reality, and UWP applications built with CUDA C/C++, OpenCL, DirectCompute, Direct3D, Vulkan, OpenGL, OpenVR, and the Oculus SDK. Check out the OpenGL frame debugging, the new Range Profiler for instant GPU bottleneck analysis, and all the new features.
Codeplay announces SPIR-V support for ComputeCpp in v0.3.0. This beta implementation of SPIR-V for OpenCL support means that developers can use SYCL and ComputeCpp to develop for any OpenCL hardware that includes a driver that consumes SPIR-V.
The landscape of APIs for accelerating vision and neural network software using specialized processors continues to rapidly evolve. Many industry-standard APIs, such as OpenCL and OpenVX, are being upgraded to increasingly focus on deep learning, and the industry is rapidly adopting the new generation of low-level, explicit GPU APIs, such as Vulkan, that tightly integrate graphics and compute. Neil Trevett presented the "Vision Acceleration API Landscape: Options and Trade-offs" tutorial at the May 2017 Embedded Vision Summit.
The Khronos Group held their annual BOF-Blitz at SIGGRAPH today. There were five BOFs in all, and they were all a huge success. If you were not able to get to SIGGRAPH and you missed the live stream, you can now watch the video online here.
Amazon AppStream 2.0 is introducing Graphics Desktop and Graphics Pro instance families to deliver high performance graphics applications from AWS. The Graphics Desktop instance family offers a single instance type with an NVIDIA GPU based on K520 with 1,536 CUDA cores, 8 vCPUs, 15 GiB system memory, and 4 GiB graphics memory. This instance type is ideal for running desktop graphics applications such as Siemens NX, SolidWorks, ESRI ArcGIS, and other applications that use DirectX, OpenGL, OpenCL, and CUDA.
ArrayFire announced the release of ArrayFire v3.5, an open source library of parallel computing functions supporting CUDA, OpenCL, and CPU devices. This new version of ArrayFire improves features and performance for applications in machine learning, computer vision, signal processing, statistics, finance, and more. Release notes are available and the source code can be found on Github.