OpenCL momentum grows
November 19, 2010 by John Taylor
Parodying the words of Dan Patrick: OpenCL is en fuego. In the last several days, both Intel and ARM made major OpenCL commitments. This Parallel Processing-themed party just got interesting.
In August 2009, AMD was the first company to submit a certification for an OpenCLTM SDK for x86 multi-core CPUs. This was a small but important step toward providing developers the necessary tools to harness this important programming environment across platforms. But we knew this was just one incremental step in our much larger and impactful strategy to drive power-efficient parallel processing forward.
In October of 2009, we again were the first to extend our SDK to include GPU support for OpenCL, and we now offer full OpenCL 1.1 support as part of our Stream SDK v2.2. In fact, we remain the only company with OpenCL support for both x86 CPUs and GPUs. But equally as important, we have a dedicated OpenCL information portal for developers to find more information and to help guide them along in the process – OpenCL Zone. Supporting both CPU and GPU is of course central to how software developers are accessing the full compute power of our new AMD Fusion Family of APUs.
When it comes to fostering true mainstream adoption of OpenCL, we understand this is a marathon, not a sprint. And, it is always easier to finish the marathon at a brisk pace when you have others running with you.
To date, AMD has worked closely with the Khronos group to help cultivate OpenCL, but we’ve been pleased to see both Intel and ARM embrace OpenCL in recent weeks. Specifically, Intel has released an alpha OpenCL SDK for Intel Core processors, while ARM introduced an embedded graphics chip with OpenCL support. In fact, ARM went so far as to say OpenCL will be on every smartphone by 2014. And, quite frankly, we couldn’t be happier than to see this happen.
As an industry, we are at an important inflection point. The advances in computer hardware have been incredible – particularly in GPUs. Look no further than the launch of our latest AMD RadeonTM HD 6800 series GPUs and the imminent launch of our AMD Fusion Family of APUs. Yet, we have only scratched the surface for what these processors can do when it comes to software. They are not just for gaming. When we can apply far more powerful, and power-efficient, GPUs to general purpose computing applications – everything ranging from financial modeling to visual applications like facial recognition – your PC experience fundamentally transforms. It becomes faster, richer and more engaging.
How do we get there? Open standards such as OpenCL.
While AMD can acknowledge that proprietary solutions helped pioneer GPGPU (including CUDA and Brook+, a standard which AMD supported for years), history has proven that open solutions tend to benefit the industry more in the long run. And in order for an open solution to really be successful, it requires broad industry support. You could say that AMD took a risk being the first processor vendor to truly embrace OpenCL and offer tools for both x86 CPUs and GPUs – but today it’s clear it was a risk worth taking as Intel, ARM and others join us. Let the parallel processing party begin.
How have you begun using OpenCL and where do you see it going from here?
John Taylor is a Director of Product Marketing at AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites, and references to third party trademarks, are provided for convenience and illustrative purposes only. Unless explicitly stated, AMD is not responsible for the contents of such links, and no third party endorsement of AMD or any of its products is implied.