Author Archive: gstoner

AFDS Was Great Turn Out for Launch of the HSA Foundation

Posted on by

We had a great turn out for the launch of HSA Foundation.  Also the HSA Foundation received some amazing feedback  from the press and analyst around the formation of HSA Foundation.  Most importantly we have had a very positive reception from industry and developer we really working on right problems and solving critical issue in today’s system..
Tom Malloy’s presentation really set up the day  with the need to address key issue in the system, drivers, and runtime to make it easier for broader class of programers to be able to solve the real problem and not get encumbared in bit banging the system for maximum performance. Phil Roger went back this year to look at real example where HSA can truely bring new levels of performance as well as performance per joule efficiency.
It is great to know  we are helping to bring better products to market that drive new levels of experience.  With HSA Foundation partners it will be ready from Deeply embedded systems, Smartphones,Tablets,Laptop, Desktops and Servers,  and will even scale all the way up to  HPC class Exaflop clusters.

  • The IP catalog they have available with the five founders is borderline staggering, and it will only get bigger. If the HSA Foundation hasn’t gotten a critical mass of coder interest too, it will soon. From here, it looks like AMD did the right thing for the right reason, don’t underestimate this one. – Charlie Demerjian, SemiAccurate
  • We have much more the learn about the HSA Foundation and its direction for the industry but we can easily say that this is probably the most important processor company collaboration announcement in many years. – Ryan Shrout, PC Perspective

Foundation Blasts Off

Posted on by

 

HSA Foundation is now in flight and already locked on target.

One year ago Phil Roger stood up at AFDS and stated he wanted to change 30 years of PC Architecture legacy,  truly address the core chalanges in getting heterogeneous paralallel computing to be approachable by all programers.  On top of this he told the Audience AMD wanted to make the specification for this open.
Roll one year forward, AMD worked with ARM, Imagination Technologies, MediaTek, and Texas Instruments  to establish the HSA Foundation.  We are all interested in solving similar issues around moving computing platform forward in mobile devices, Tablets, Laptops, Workstation, and Servers.   On the anniversary of 2011 AFDS, the team has achieved it goal of bring your truly open specifications around HSA technology via the HSA foundation. Which has very solid team  of passionate people and founding companies who will drive this capabilities forward into market.

We look forward to working with all of you around this exciting new initiative.

HSA Foundation Start Of New Era In Computing

Posted on by

AMD, ARM, Imagination, MediaTek and Texas Instruments Unleash the Next Era of Computing Innovation

–Technology Leaders Establish the HSA Foundation to Create a Unified, Open Industry Standard Architecture for Heterogeneous Processing; Heterogeneous Processors, Which Combine Two or More Different Types of Processors into a Single Chip, Accounted for Nearly Half of All Microprocessor Revenue in 20111

BELLEVUE, WASH. – June 12, 2012 – Today at the AMD Fusion Developer Summit (AFDS), global technology leaders came together to announce the formation of the Heterogeneous System Architecture (HSA) Foundation The HSA Foundation is a non-profit consortium established to define and promote an open, standards-based approach to heterogeneous computing that will provide a common hardware specification and broad support ecosystem to make it easier for software developers to deliver innovative applications that can take greater advantage of today’s modern processors.
AMD (NYSE: AMD), ARM, Imagination Technologies, MediaTek Inc., and Texas Instruments (TI) (NYSE: TXN) are the initial founding members of the HSA Foundation. The companies will work together to drive a single architecture specification and simplify the programming model to help software developers take greater advantage of the capabilities found in modern central processing units (CPUs) and graphics processing units (GPUs), and unlock the performance and power efficiency of the parallel computing engines found in heterogeneous processors.
By standardizing the heterogeneous programming model, developers can more easily and cost-effectively develop new software tailored to take advantage of the expanding $55.5B heterogeneous (also known as “hybrid”) processor market1. The HSA Foundation will help usher in the next era of software innovation, from enabling unprecedented user experiences to improving cloud-based data management, streaming, and security.
“HSA moves the industry beyond the constraints of the legacy system architecture of the past 25-plus years that is now stifling software innovations,” said Phil Rogers, HSA Foundation President and AMD Corporate Fellow. “By aiming HSA squarely at the needs of the software developer, we have designed a common hardware platform for high performance, energy efficient solutions. HSA is unlocking a new realm of possibilities across PCs, smartphones, tablets and ultrathin notebooks, as well as the innovative supercomputers and cloud services that define the modern computing experience.”
As an independent consortium, the HSA Foundation is open to any and all computing industry professionals with an interest in driving the next era in computing performance and energy efficiency. The HSA Foundation welcomes forward-thinking semiconductor companies, platform and OS vendors, device manufacturers, independent software vendors, academia and open source developers. Members of the HSA Foundation plan to deliver robust development solutions for heterogeneous compute to drive innovative content and applications with developer tools, software developer kits (SDKs), libraries, documentation, training, support and more.
“One year ago, AMD boldly announced a roadmap for making HSA a reality, starting with combining the CPU and GPU as a unified processing engine to create an architecture that would greatly simplify the programming of heterogeneous platforms,” said Manju Hegde, corporate vice president of Heterogeneous Applications and Developer Solutions for AMD. “Today, we are continuing our heterogeneous computing leadership and dedication to driving broad industry adoption by making the HSA architecture an open standard.”
“We are all demanding more from the technology that connects us to our digital worlds. Graphical interfaces are critical to the user experience but can have a power impact. With open standards, developers can now provide outstanding graphics without compromising power-efficiency,” said Jem Davies, ARM Fellow and vice president of Technology, Media Processor Division, ARM. “ARM welcomes the formation of the HSA Foundation and is pleased to be one of the founding members. ARM’s extensive experience with heterogeneous systems brings a unique leadership perspective to developing the right compute processor for the right tasks based on the latest ARM technology, such as ARM® Mali™ GPUs and Cortex™ processors.”
“Imagination is delighted to be one of the founders of the HSA Foundation. Our long-held view is that heterogeneous processing architectures are fundamental to scalability, flexibility and performance for future SoC (Systems-on-Chip) design,” said Tony King-Smith, vice president of Marketing, Imagination Technologies. “We believe that the combination of CPUs with increasingly powerful on-chip GPUs, using industry standard APIs such as OpenCL™, Renderscript Compute, and Direct Compute, is the vital first step along this path. The HSA APIs and tools complement these higher level APIs to help SoC systems engineers create future generation computing platforms. As applications developers embrace the benefits of embracing such architectures, the HSA Foundation will provide a valuable cross-platform, cross-OS means of ensuring maximum heterogeneous application portability and optimization. We look forward to working closely with all of the founders and other members of the HSA Foundation to deliver this key open standard that is set to fuel the future of mass market high performance computing.”
“MediaTek is excited about the potential of heterogeneous computing in enabling innovative mobile applications. We believe that an open standard is key to bringing the technology to the mainstream smartphone market, said Chien-Ping Lu, senior director, Corporate Technology Office, MediaTek. “As a chipset and platform supplier with a strong focus on mainstream smartphones, we are excited to join AMD, our IP partners, Imagination, ARM, TI and other industry leaders to promote HSA as the open standard for heterogeneous computing.”
“TI’s expertise in delivering low-power, high-performance experiences via the smart, multicore architecture of our OMAP™ platform undoubtedly makes our position as a founding member of the HSA Foundation a natural fit,” said Matthew Locke, director, Linux Development Center, TI. “We look forward to working with other leaders in the industry to drive adoption of heterogeneous computing as a standard platform across embedded, mobile, personal and cloud computing.”
In addition to the founding members, contributor and supporting-level members will assist with research, development, production, manufacture, use, and the sale of HSA and heterogeneous computing software. For more information about the HSA Foundation’s membership levels and benefits or to join the organization, please go to www.hsafoundation.com.
Supporting Resources

  • For additional information about HSA and the HSA Foundation visit the HSA Foundation website at www.hsafoundation.com.
  • Follow the HSA Foundation on Twitter: @HSAFoundation.
  • “Like” the HSA Foundation on Facebook: www.facebook.com/HSAFoundation.
  • View Phil Rogers’ AFDS keynote presentation via AFDS-D. Register here.

About Heterogeneous System Architecture (HSA)
Developers will benefit from the open standard programming of HSA for both the CPU and GPU, which allows the two processors to work cooperatively and directly in system memory. Additionally, HSA provides a single architecture across multiple operating systems and hardware designs. By maximizing the full compute capabilities of systems with both CPUs and GPUs, users can see performance and energy efficiency boosts across a variety of applications.

About the HSA Foundation  
The HSA (Heterogeneous System Architecture) Foundation is a not-for-profit consortium for SoC IP vendors, OEMs, academia, SoC vendors, OSVs and ISVs whose goal is to make it easy to program for parallel computing. HSA members are building a heterogeneous compute ecosystem, rooted in industry standards, for combining scalar processing on the CPU with parallel processing on the GPU while enabling high bandwidth access to memory and high application performance at low power consumption. HSA defines interfaces for parallel computation utilizing CPU, GPU and other programmable and fixed function devices, and support for a diverse set of high-level programming languages, thereby creating the next foundation in general purpose computing.
About AMD
AMD (NYSE: AMD) is a semiconductor design innovator leading the next era of vivid digital experiences with its groundbreaking AMD Accelerated Processing Units (APUs) that power a wide range of computing devices. AMD’s server computing products are focused on driving industry-leading Cloud computing and virtualization environments. AMD’s superior graphics technologies are found in a variety of solutions ranging from game consoles, PCs to supercomputers. For more information, visit http://www.amd.com.
AMD, the AMD Arrow logo, , and combinations thereof, are trademarks of Advanced Micro Devices, Inc. Other names are for informational purposes only and may be trademarks of their respective ownersThe HSA Foundation logo is a trademark of the Heterogeneous System Architecture Foundation.
1Half of $111B 2011 processor revenue generated by hybrids, according to IMS Research, recently acquired by IHS. http://imsresearch.com/press-release/Half_of_111_Billion_2011_Processor_Revenue_Generated_by_Hybrids_According_to_IMS_Research

HSA Brings GPGPU Computing and Extends It Further

Posted on by
GPU are great at attacking problems that come in the following form.
•Data Parallelism – SPMD
•Embarrassingly Parallel Application
–Rendering – Raster based
–Simulation of large scale particle system
–Large Graphs/networks
–Numerical Integration
–Monte Carlo Integration
–Bioinformatics
–Genetic Algorithms and evolutionary computation metaheuristics
–Ensemble Calculations
–Visual Mining – ex Large Scale Face Recognition
–Brute force searches in cryptography
–Distributed Set Processing  of relational data
–Sieve Analysis for GNFS and QS for integer factorization
HSA also bring about wider set of problem by addressing
•What GPU do well today
–Data Parallelism – Embarrassingly Parallel Application
–SPMD
•Plus a much richer set of Parallel Solutions
–Task-Parallelism,
–Nested-Parallelism,
–Braided-Parallelism,
–Irregular parallelism
–MPMD
•Parallel Problems that are Communication Intensive
–Need High Bandwidth Low Latency Interconnects
–Examples
•Solution that are Parallelizable by Domain Decomposition
•Partial differential equations on a regular grid using discrete time stepping
–More efficient implementation of MapReduce, Hash Tables, Sparse Matrix Vectors and Conjugate Gradients solvers, FETI-DP Method, List Ranking, and spatial search
•Algorithm that need Inter-task Communication
•Application with the need of branching support.
•Application that need exceptions processing
•Dynamic load balancing task between processing elements
What is revolutonary
•What GPU do well today
–Data Parallelism – Embarrassingly Parallel Application
–SPMD
•Plus a much richer set of Parallel Solutions
–Task-Parallelism,
–Nested-Parallelism,
–Braided-Parallelism,
–Irregular parallelism
–MPMD
•Parallel Problems that are Communication Intensive
–Need High Bandwidth Low Latency Interconnects
–Examples
•Solution that are Parallelizable by Domain Decomposition
•Partial differential equations on a regular grid using discrete time stepping
–More efficient implementation of MapReduce, Hash Tables, Sparse Matrix Vectors and Conjugate Gradients solvers, FETI-DP Method, List Ranking, and spatial search
•Algorithm that need Inter-task Communication
•Application with the need of branching support.
•Application that need exceptions processing
•Dynamic load balancing task between processing elements

ARM Mali GPUs at the AMD Fusion Developer Summit

Posted on by

I am excited to be speaking at the AMD Fusion Developer Summit. I shall be speaking, but in a talk that is not listed as part of the agenda, but as part of one of the keynote talks. As many of you may know, my appearance last year generated a lot of speculation about the nature of the relationship between ARM and AMD. I gave a talk about the things that the two companies agreed on rather than what we disagreed on. Mostly I talked about OpenCL and the importance of open standards going forward, and how that related to heterogeneous compute systems.
This year, we have a great deal to discuss. ARM is all about low power and many people in the industry now realize that GPUs have a central role to play in providing highly energy-efficient computing. It’s an exciting future that can grow the ecosystem that surrounds computing. ARM’s unique portfolio of CPU, GPU, interconnect and physical IP puts us at the forefront of one of the most important technological changes in a long time. Reflecting on that and some of those changes, I will be making an announcement at the show.
I wonder what we will do together as a result of these momentous changes in the industry? It promises to be an exciting year ahead as the ecosystem will be strengthened even further.
Leave a comment on this blog. Tweet me @ARMMultimedia. Leave a note on the ARM Facebook page with your questions for me?
Jem is an ARM Fellow and likes to think of himself as “The Godfather” to technical talent in ARM. After spending some time in his youth writing software for satellites and traffic-lights among other fascinating things, Jem spotted the technical inflection point of the mobile industry: graphics, video and other visual computing. As VP of technology in the Media Processing Division of ARM, Jem is busy with a lot of projects involving the future of cool ARM technology, which will revolutionise how people experience and interact with digital devices.

AFDS: Key Presentations To Go Deeper Into What Is HSA

Posted on by

AFDS Keynotes

  • Phil Rogers, corporate fellow, AMD: “The Programmer’s Guide to a Universe of Possibility”
  • Tom Malloy, senior vice president & chief software architect, Advanced Technology Labs, Adobe Systems Incorporated: “The Promise of Parallel: Today’s State of Heterogeneous Computing”
  • Dr. Amr Awadallah, co-founder and CTO, Cloudera: “Apache Hadoop: The Modern Data Operating System”
  • Mark Papermaster, senior vice president and chief technology officer, AMD: “Heterogeneous Computing: The Market. The Roadmap”

AFDS Presentations

  • IOMMUv2: The Ins and Outs of the Heterogeneous GPU Use
  • GPGPU algorithms:  How Heterogeneous System Architecture can be leveraged to optimize such algorithms in video games
  • Accelerating SURF on HSA
  • Bolt: A C++ Template Library for HSA
  • An overview of HSAIL
  • HSA From A HPC Usage Perspective
  • HSA Memory and Execution Model à la HSA Runtime