Khronos Demonstrates OpenCL Momentum at SC09
OpenCL Conformance Tests Now Available With Conformant Products Shipping;
Working Group Membership Expands to Thirty-Three High-Performance Computing, Gaming, Middleware, System and Silicon Vendors
November 16, 2009 – SC09 Conference (Portland OR) – The Khronos™ Group, an industry consortium with more than 100 members working together to create open standards for authoring, accelerating and accessing visual computing, today announced strong industry support and a wide range of shipping products utilizing OpenCL™ at SC09 (an international conference for high-performance computing). A number of demonstrations and tutorials from Khronos Group members highlighting the power and scalability of OpenCL will also be a major part of SC09.
OpenCL is the open, royalty-free standard for general-purpose parallel programming across CPUs, GPUs and other processors. OpenCL provides software developers portable and efficient access to the full power of a wide range of systems including high-performance compute servers, desktop computer systems and handheld devices. The OpenCL 1.0 specification and more details are available at www.khronos.org/opencl/.
OpenCL Conformant Products Shipping
Khronos released a conformance suite for the OpenCL 1.0 specification in May 2009 that exhaustively tests both the functionality and numerical accuracy of OpenCL implementations before they are licensed to use the OpenCL trademark. A number of shipping products from working group members have successfully passed OpenCL conformance including products from AMD, Apple and NVIDIA. The current list of conformant products can be found at: http://www.khronos.org/adopters/conformant-products/#topencl
“By enabling cross-platform development for heterogeneous architectures, OpenCL is helping to bring GPU compute capability to mainstream applications,” said Patricia Harrell, director of stream computing at AMD. “AMD fully supports the advantages of industry standard development through its ATI Stream SDK v2.0, an OpenCL 1.0 compliant solution for both ATI GPUs and x86 CPUs.”
“NVIDIA cares deeply about ensuring that OpenCL developers have the tools they need to easily develop and deploy mainstream applications on more than 150 million NVIDIA OpenCL 1.0-capable GPUs,” said Sanford Russell, general manager, GPU Computing software at NVIDIA. “Our latest R195 driver provides the industry’s most complete support for OpenCL 1.0 with NVIDIA’s OpenCL Visual Profiler and new OpenCL extensions including double precision, ICD and imaging. We are very excited about the increased momentum that OpenCL is creating behind GPU Computing.”
OpenCL Working Group Experiences Significant Membership Growth
The membership of the Khronos OpenCL working group has grown steadily since the release of the OpenCL 1.0 specification, with a rich diversity of companies helping to evolve and support the specification. Membership in the OpenCL working group now includes thirty-three market leaders: Activision Blizzard, AMD, Apple, ARM, Broadcom, Codeplay, Electronic Arts, Ericsson, Freescale, Fujitsu, GE, Graphic Remedy, HI, IBM, Intel, Imagination Technologies, Los Alamos National Laboratory, Motorola, Movidius, Nokia, NVIDIA, Petapath, Presagis, QNX, Qualcomm, S3, Samsung, STMicroelectronics, Takumi, Texas Instruments Incorporated (TI), Toshiba, Vivante and ZiiLABS (formerly 3DLABS).
“Codeplay is very excited about OpenCL because it enables us to adapt our development tools to allow developers to accelerate their C++ software on any device that supports OpenCL,” said Andrew Richards, CEO of Codeplay. “By providing a standard acceleration API, OpenCL enables software developers to invest in GPGPU with confidence that their investment is future-proof.”
“IBM is encouraged by the rapid acceptance of support of the Khronos Group’s OpenCL 1.0 specification, and is pleased to announce the availability of the OpenCL Development Kit for Linux on Power on AlphaWorks,” said Chris Maher, vice president HPC Development at IBM. “IBM recognizes the significance of a portable, high performance, heterogeneous programming language and with this technology preview will enable programmers to easily explore these new programming paradigms on Power and Cell/B.E. Processors.”
“We’ve been working with several of our key licensees and application middleware partners studying advanced usage of our GP-GPU technology,” said Tony King-Smith, Imagination Technologies. “Results so far show that POWERVR SGX graphics cores already in volume production achieve significantly better performance for some algorithms than traditional CPUs clocked 5 times faster when using GLSL under OpenGL ES. Thanks to our highly efficient, scalable unified shader architecture, we can see ways to achieve far more performance using OpenCL with our latest SGX Series5XT cores. We are getting strong feedback from both semiconductor and OEM partners that they wish to use OpenCL as the means of unlocking GP-GPU capabilities in a wide range of applications in mobile, automotive and home entertainment, and Imagination is committed to providing fully compliant OpenCL API support across a significant number of our POWERVR SGX IP cores.”
“Our development experience at Los Alamos National Laboratory with adapting codes to the Roadrunner supercomputing architecture exposed several areas where there were no obvious tools or techniques that would allow us to maintain portability across the variety of platforms that we must routinely support to fulfill the Laboratory’s stewardship mission,” said Ben Bergen, Evolving Applications and Architectures Team, Los Alamos National Laboratory. “Initial proof-of-concept experiments with the OpenCL framework make us optimistic that OpenCL can address many of the challenges that we will be facing as the HPC landscape evolves into the future.”
“TI applauds these OpenCL milestones, as they reiterate the opportunities this heterogeneous standard brings to multiple markets,” said Ameet Suri, marketing manager, OMAP product line, wireless business unit, TI. “OpenCL’s portable, cross platform and high-performance framework pairs with TI’s OMAP applications processors to create a new world of mobile user experiences, and ensures that developers have full access to the parallel and heterogeneous compute capabilities like multiple CPU, GPU, DSP and other imaging accelerators provided by the underlying OMAP platform that will shape the future.”
“Many of our licensees are looking to maximize the potential of embedded GPUs for demanding computing tasks that will differentiate their SoC solutions,” said Wei-Jin Dai, president and CEO of Vivante Corporation. “Vivante is providing native support for OpenCL as a driver for unlocking the power of our GPUs as a compute element in a wide variety of embedded applications.”
OpenCL Sessions at SC09:
The Khronos Group is pleased to participate at SC09 in trade show booth #242 and at multiple OpenCL sessions:
Tutorial : OpenCL - A Standard Platform for Programming Heterogeneous Parallel Computers
- Date: Monday, 16 November 2009, 1:30PM – 5:00PM
- Tim Mattson - Intel,
- Ian Buck - NVIDIA, Mike Houston and Ben Gaster – AMD
BOF (Birds of a Feather) : Can OpenCL Save HPC?
- Date: Wednesday, 18 November 2009 5:30PM – 7:00PM
- Speakers: Tim Mattson – Intel, Ben Gaster – AMD, John Stone - University of Illinois at Urbana-Champaign and Marcus Daniels - Los Alamos National Laboratory
- Moderator: Ben Bergen, Computational Physics, Los Alamos National Laboratory
More details of OpenCL sessions at SC09 can be here.
Visitors to the Khronos booth or the OpenCL sessions at SC09 will be offered a free laminated OpenCL Quick Reference card that covers all important aspects of the OpenCL API. The OpenCL Quick Reference Card is also available online.
OpenCL defines a high-performance, portable parallel programming abstraction to accelerate a wide range of applications, including consumer, media, scientific and HPC solutions. By creating an efficient, close-to-the-metal programming interface, OpenCL is the foundation layer of a parallel computing ecosystem of platform-independent tools, middleware and applications. OpenCL consists of an API for coordinating parallel computation across heterogeneous processors and a cross-platform kernel programming language with a well-specified computation environment.
About The Khronos Group
The Khronos Group is an industry consortium creating open standards for authoring, accelerating and accessing visual computing. Khronos standards include OpenGL®, OpenGL® ES, WebGL™, OpenCL™, OpenMAX™, OpenVG™, OpenSL ES™, OpenKODE™, OpenWF™ and COLLADA™. All Khronos members are able to contribute to the development of Khronos specifications, are empowered to vote at various stages before public deployment and are able to accelerate the delivery of their cutting-edge media platforms and applications through early access to specification drafts and conformance tests. More information is available at www.khronos.org.
Khronos, WebGL, OpenKODE, OpenVG, OpenMAX and OpenWF are trademarks of the Khronos Group Inc. OpenCL is a trademark of Apple Inc., COLLADA is a trademark of Sony Computer Entertainment Inc. and OpenGL is a registered trademark of Silicon Graphics Inc. used under license by Khronos. All other product names, trademarks, and/or company names are used solely for identification and belong to their respective owners.
Khronos Group Launches OpenWF Standard to Accelerate Windowing on Mobile and Embedded Devices
Composition and Display Layering Functionality for Accelerated Windowing Systems;
Integration with Khronos APIs such as OpenGL ES, OpenMAX and OpenVG
November 9, 2009 – Clearlake Park, CA - The Khronos Group, an industry consortium with over 100 members working together to create open standards for authoring, accelerating and accessing visual computing, today announced the release of the OpenWF™ 1.0 standard. OpenWF is an operating system-independent and hardware-neutral foundation for building windowing systems and providing display control functionality in accelerated mobile and embedded devices.
OpenWF enables a new degree of portability, acceleration and abstraction for windowing systems, while adding functionality and features through close integration with Khronos application APIs. OpenWF acts as the underlying route to the display for advanced graphics and multimedia content generated using APIs such as OpenGL® ES for fast and portable 3D graphics, OpenVG™ for vector graphics acceleration, and OpenMAX™ for multimedia.
OpenWF provides two separate but complementary low-level APIs for composition of content and the configuration of display devices: OpenWF Composition and OpenWF Display. OpenWF Display enables portable access to display control hardware for manipulating screen attributes, while OpenWF Composition allows for layering and system-wide composition of application content. The two APIs can be used together or independently, depending on specific platform needs.
OpenWF enables highly-optimized mechanisms to display application content in a windowing system. By eliminating redundant memory accesses, consumed memory bandwidth may be reduced by more than half, resulting in significant power savings. Specific details on features and benefits of OpenWF may be found at http://www.khronos.org/openwf.
Wide range of industry leaders express strong support for OpenWF:
“ARM is looking forward to the wide adoption of OpenWF across multiple operating systems,” said Ian Smythe, director of marketing, Media Processing Division of ARM. “The opportunity for a standard cross platform composition/display API is one we believe will benefit ARM's partners and speed the introduction of high efficiency video and graphics functionality into a wide range of partner products.”
“OpenWF is a valuable extension to the family of Khronos standards,” said Mark Casey, vice president and general manager of mobile multimedia at Broadcom. “These standards help deliver a dramatic improvement in mobile multimedia user experience by simplifying the integration of low-power hardware accelerated architectures into mobile devices.”
“If the UI is slow, nobody cares which CPU or GPU is inside the device,” said Petri Kero, CTO of drawElements Ltd. “System software needs to take full advantage of whatever is under the hood. We have been working on this for a long time already, and we're very excited to see OpenWF standardize the key interfaces.”
“OpenWF is a key graphics technology for Nokia, allowing us to quickly take advantage of diverse hardware platforms,” said Robert Palmer, graphics technology architect at Nokia.
“OpenWF provides a critical missing link for rich media content presentation pipeline. It rids the industry of proprietary methods of uniting media sources with displays and results in faster time to market and better utilization of hardware,” said Jarkko Kemppainen, director, Platform Business at Symbio.
“OpenWF is a system-level API for use by operating system vendors that complements the Khronos mobile APIs for application development,” said Neil Trevett, president of the Khronos Group. “OpenWF enables low-cost, hardware-accelerated 2D composition - providing an alternative to OpenGL ES-based composition using 3D-intensive GPUs. OpenWF allows platform integrators a wider choice of user interface, price and functionality options within the Khronos ecosystem.”
Key features of OpenWF:
- Provides layered 2D composition functionality implementable on low-power hardware
- Enables system-wide composition of application content
- Includes alpha-blending, masking, resizing, rotation, mirroring
- Defines scalable acceleration for low-end dedicated hardware through to high-end GPUs
- Optimizes use-cases, such as long-running video playback with subtitles, for low power
- Defines a windowing system-level interface to control display hardware
- Provides configuration control over mode-setting of built-in and external displays
- Includes discovery, power, resolution, rotation and pipeline control
- Supports multiple interfaces including HDMI, DVI, S-Video, embedded LCD panels and legacy display adapters
About The Khronos Group
The Khronos Group is an industry consortium creating open standards for authoring, accelerating and accessing visual computing. Khronos standards include OpenGL®, OpenGL® ES, OpenCL™, OpenMAX™, OpenVG™, OpenSL ES™, OpenKODE™, OpenWF™ and COLLADA™. All Khronos members are able to contribute to the development of Khronos specifications, are empowered to vote at various stages before public deployment and are able to accelerate the delivery of their cutting-edge media platforms and applications through early access to specification drafts and conformance tests. More information is available at www.khronos.org.
Khronos, OpenKODE, OpenVG, OpenSL ES, OpenMAX and OpenWF are trademarks of the Khronos Group Inc. OpenCL is a trademark of Apple Inc., COLLADA is a trademark of Sony Computer Entertainment Inc. and OpenGL is a registered trademark and the OpenGL ES and OpenGL SC logos are trademarks of Silicon Graphics International used under license by Khronos. All other product names, trademarks, and/or company names are used solely for identification and belong to their respective owners.