Skip navigation
1 2 3 Previous Next

AMD Business

45 Posts authored by: brystal.boyd Employee

Chappie Blog Image_hpc.pngYou may be surprised to know that whatever you are using to read this - a notebook, desktop, tablet or smartphone - can be part of a supercomputer.

 

In the hit film “Chappie” about a futuristic robot, audiences are taken to a world where technology is used in exciting and unexpected ways. In a scenario that put a smile on the face of AMDers worldwide, in one scene a number of Sony PlayStation 4 consoles, each with AMD Radeon™ graphics and AMD CPU processing inside, are connected together, effectively combining into a supercomputer cluster. While “Chappie” is a Hollywood creation, the idea behind connecting game consoles to create a supercomputer is not as science fiction as we may think. Clustering allows almost any device to become part of a supercomputer by being part of a pool of resources.

 

The world’s first supercomputers used expensive, specialized processors that put high performance computing out of reach most organizations, but today, many of the most powerful supercomputers make use of off-the-shelf processors to produce “nodes.” These nodes are then connected together to form clusters, similar to what we see with an array of AMD-powered PS4s in “Chappie”.

 

Supercomputer clusters typically make use of server-orientated processors such as the AMD Opteron™ family of processors, but occasionally academics and engineers want to try something different and use less specialized computing resources. Because combining nodes to form a cluster is done using widely available equipment such as Ethernet, it is possible to make a cluster from a number of other devices, such as the Gizmo 2 or game consoles, for example.

 

To create the supercomputer cluster in “Chappie”, a number of PlayStation 4 consoles, each packing 1.84 teraflops of AMD computing power inside are connected together.  That processing power is delivered in a System-on-Chip design with eight AMD CPU cores and 18 AMD Radeon Graphics Core Next (GCN) compute units, typically used to process all the software, games, videos, and motion sensor capabilities that enable players to interact with others through online services. Even more impressive, the same compute architectures found in the PlayStation 4 really are powering some of today’s supercomputer clusters.

 

In recent years, high performance compute clusters achieved their immense compute capabilities through a mix of processor technologies. The CPU, which has been at the heart of supercomputers for decades has been supplemented with accelerators such as AMD FirePro GPUs, which provide immense compute capability and do so in an energy efficient manner.

 

Energy efficiency is a key to making a powerful high performance compute cluster given cooling thousands of processors is both a budgetary and engineering challenge. That is why power efficient CPUs and GPUs make ideal components to build compute clusters and have been a significant contributing factor in the massive growth in the amount of compute power in supercomputer clusters. Such is the efficiency of GPUs, the world’s most energy efficient cluster makes use of AMD FirePro accelerators.

 

By pooling compute and storage resources, academics, scientists, corporations and governments are able to tackle some of the world’s most challenging questions, from medical research to mineral exploration. In many cases, supercomputer clusters are made up of more than 10,000 nodes, with more than 100,000 compute cores, each core working towards producing a final result.  And if you think that only a scientist gets to feel the effects of a supercomputer cluster you may be surprised. When you glance at the weather forecast in the morning you may be surprised to find out that it is in part calculated by running immensely complex forecasting models on clusters. Cool, right?  Now think about this; supercomputers do even more than just science and predicting the weather. The car you drive has most likely been partly designed and tested using a supercomputer that is analyzing computational fluid dynamics to increase fuel efficiency.

 

One might say that supercomputer clusters have played a significant part in the world around us and with the growth in machine learning and big data analysis; it is set to play an even bigger role in our daily lives in the future. Perhaps the most amazing part in all this is that supercomputer clusters are powered by the same type of technology that is found under your desk or television and, you probably didn’t even realize how powerful these common devices actually can be.

 

John Taylor, is the Corporate Vice President of Marketing at AMD. His/her 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. There is no endorsement, express or implied, of this blog by the movie “Chappie”. 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.


*Originally Posted by llatif in AMD Business on Apr 28, 2015 1:56:37 PM

The AMD FirePro™ graphics team, working closely with PTC®, enables engineering efforts to ensure exceptional workflow performance for PTC Creo® Parametric 3.0 users, giving them access to features and performance that open the door to breakthrough innovation and outstanding competitive advantages.

 

motocycle.jpg

Robustness and scalability for large assemblies – rapid design and greater “design intuition”

Developed by AMD in close collaboration with PTC, advanced OpenGL optimizations greatly increase the 3D frame rate and interactivity of PTC Creo. The new transparency mode exploits the on-board memory processing of modern GPUs to enable additional, always-on 3D acceleration on complicated datasets and complex workflows. As a result, PTC Creo designers using AMD FirePro graphics experience up to 40x faster 3D frame rates than PTC Pro/ENGINEER Wildfire 5(1), helping them maintain their productivity and creative flow.

 

Reliability and graphics optimization – tuned, optimized, and certified for PTC Creo

To ensure optimal performance and compatibility, AMD FirePro professional graphics solutions are tested and certified by PTC across the suite of applications. Combined with workstations certified by PTC, AMD FirePro professional graphics provide an unbeaten user experience for PTC Creo designers with up to 2.5x faster 3D frame rates in shaded mode(2), compared to competitive GPU solutions.


Collaboration and communication across workflows - Creo 3.0 Unite technology and AMD FirePro plus AMD Eyefinity technology
Product development workflows have changed significantly over recent years. Working with multiple applications is common with design, simulation, data management and collaboration all happening in unison. AMD FirePro graphics cards feature AMD Eyefinity(3) multi-display technology that empowers engineers to view multiple applications and product assemblies across three, four or even six high-resolution monitors all from a single graphics card, at up to 4K resolution for each output. Users can view designs at ultra-high resolutions for increased design accuracy, realism and enhanced insight, and speed up workflow by using the extra screens to view additional applications. Using the new Creo 3.0 Unite Technology and AMD Eyefinity, collaboration and multi-CAD data consolidation across the family of PTC products in design, manufacturing and simulation can be accomplished with ease.

 

Improved graphics realism and performance – Creo 3.0 new immersive modeling environment

Fully supported by AMD FirePro graphics, Creo 3.0 now allows interactive and realistic design visualization. Advanced features such as reflections, HDR, and ambient occlusion in an immersive environment allows designers to experience the product, not just visualize it. AMD recommends the following professional graphics cards for PTC Creo users:

 

AMD FirePro™ W5100 is recommended for design and simulation,

AMD FirePro™ W7100 is recommended for large assemblies and rendering.

 

To learn more about these technologies, visit http://www.fireprographics.com/ws/cad/creo/index.asp

 

Antoine Reymond is an industry executive for Professional Graphics at AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied.

 

1 Based on comparison of AMD FirePro W7100, Creo 3.0 in OIT transparency mode vs. Pro/ENGINEER Wildfire 5.0 blended transparency mode, running AMD internal benchmark “Creo Benchmark v3.1” using AMD’s Internal Motorcycle dataset. Intel Xeon X5570 at 2.93GHz 4-Core, 12GB RAM, Windows 7 64-bit, 120GB SSD, AMD FirePro driver 14.301, display 1920x1200, Creo Parametric 3.0 F000, and PTC Pro/ENGINEER Wildfire 5.0 M250.

 

2 Based on comparison of AMD FirePro W7100 vs Nvidia Quadro K4200 in Creo 3.0 running AMD internal benchmark “Creo Benchmark v3.1” using AMD’s motorcycle dataset and shaded mode. Windows 7 64-bit, Intel Xeon E5-2630 at 2.30 GHz, 18GB RAM, Seagate 250 GB 7200 SATA 16 MB HDD, AMD driver 14.30-140915a-175805E ATI, Nvidia driver 340.66.FP-114

 

3 4K content required. AMD Eyefinity technology supports up to six DisplayPort™ monitors on an enabled graphics card. Supported display quantity, type and resolution vary by model and board design; confirm specifications with manufacturer before purchase. To enable more than two displays, or multiple displays from a single output, additional hardware such as DisplayPort-ready monitors or DisplayPort 1.2 MST-enabled hubs may be required.  A maximum of two active adapters is recommended for consumer systems. 4K support requires 4K display and content.  See www.amd.com/eyefinityfaq for full details.


*Originally Posted by System Admin in AMD Business on Apr 21, 2015 8:53:29 AM

AMD PRO has been gaining momentum in the market. As a trusted provider of innovative technology solutions, AMD is inside the latest generation of console units, jetliner cockpit displays, the world’s fastest energy efficient computer and the systems on an increasing number of workplace desks.

 

So what is making AMD PRO based systems so popular? In addition to the excellent performance and value that are offered by AMD PRO A-Series APUs and the reliability offered with a longer warranty (among a host of other sought after features as you can see on the PRO website), AMD continues to develop and support industry standards. One of the industry standards supported by AMD is the Distributed Management Task Force (DMTF) Desktop and mobile Architecture for System Hardware (DASH) Standard.

In October of 2014, we released the AMPS 2.1 plug-in for SCCM 2012. The AMPS plug-in extends SCCM 2012 to manage the desktop and mobile systems' out-of-band (irrespective of the power state or the clients' operating system) using DASH standards.

 

We are pleased to announce the availability of the AMPS 2.2 release to further extend SCCM features and functionality. In the AMPS 2.2 release, additional features supported are:

 

    • Text Redirection
      • Text redirection provides BIOS-assisted console and keyboard redirection to a terminal in the SCCM console.
    • USB Redirection
      • USB Redirection can be used to boot the managed systems to a remote image file such as .iso.
    • Scheduled Power
      • Schedule (at a particular date and time) power operations on a collection of devices.
    • Alert Subscription and Reception
      • Subscribe, receive and encode DASH alerts from managed devices.
    • SCCM 2012 CAS support
      • AMPS supports DASH operations from Central Administrative Site of SCCM 2012.

 

You can find more information on AMD’s DASH manageability solutions in the “Documents and White Papers” section of the AMD Commercial Tools and Resources page. Additional documentation and downloads including AMPS 2.2 are available on the Tools for DMTF DASH page. Please reach out to your AMD representative or visit the forum with any questions or feedback you may have.

 

Jay Hiremath is a Sr. Director - Platform & Software Engineering 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.

 

*Originally Posted by System Admin in AMD Business on Mar 30, 2015 11:33:29 AM

One of the great – and terrifying – things about computing is the sheer rate of change. What is the expensive high end today, a moment later becomes the minimum necessary for reasonable performance. Both the rate of development and the ability of the technology industry to feed innovations and improvements down to lower and lower price points is stunning.

The pay-off for this, of course, is the democratization of technology. A decade ago, taking video footage and editing it to a good standard was a task for professionals. Computer-aided design (CAD) was something you did in a design studio, and virtual reality was more virtual than reality. Today however, the value PC you buy from the shop around the corner can cut, splice, add after effects and output to a quality that was previously the domain of high end post-production houses and newsrooms.

This gives rise to a legitimate question: why bother buying the latest and greatest when the mid-range is perfectly capable? I believe it’s a little more complicated than just a simple matter of price.

 

The difference between stellar experiences and mission critical

 

First, let’s establish what’s different between a consumer and professional graphics card. The average home or business use for a graphics card is not mission critical. Most business PCs, for example, might be used for standard office productivity applications – and consumer graphics cards are excellent for the job they are intended for. In fact, as time goes by, graphics cards are arguably more important than the CPU when it comes to handling quite a lot of the computational load the average user puts on their PC.

When it comes to the latest and greatest games, they are also easily handled, as graphics cards are truly keeping pace with everything the gaming industry throws at them. The bottom line is that the experience offered by consumer graphics cards today are perfect for consumers.

Professional graphics needs, however, are different. They might be used in a multi-GPU installation in a data center for processing huge chunks of information. They might be used for a CAD application tasked with building the design for the next range of high performance cars. They might be rendering the next big Hollywood blockbuster movie. All of these are what we’d call business critical applications, and all of these rely on predictable performance that works reliably every time.

 

The older the better?

 

There’s another element to this reliability story; most business-critical workstations are built from the ground up with a certain specification. Change that spec, and the whole process needs to be reworked again.

Having access to older solutions, counter-intuitive as it might sound in today’s world, is actually vital. This is one of the reasons why NASA very rarely updated any of the technology in the space shuttle: it worked, and worked reliably and predictably. When the slightest change could mean catastrophe, the focus must be on having a known solution, not a new one.

With the high costs of planning and implementation for mission critical applications, it’s more important to replace professional graphics cards with something exactly the same as opposed to different solutions that may be incompatible with the application you’re running. Therefore, having a longer warranty is vital - for example, AMD FirePro’s warranty is three years. Additionally, access to such products for a longer period of time is equally important – exemplified by AMD FirePro offering two to five years availability.

Finally, there’s the software drivers and interoperability. The makers of professional software programs are very keen to maximise the performance of their applications, and spend a great deal of time making sure the programs are optimised for professional-grade workstations and components. You can be sure, with applications like Autodesk Maya™, Adobe After Effects and AutoCAD running on an AMD FirePro graphics card, the hardware and software are built to run together.

 

 

Infographic-final.jpg

 

The consumer contrast

 

Consumer grade cards offer their own benefits and are built for specific purposes; they are priced affordably for consumers, and the technology receives regular refreshes and can be blazingly fast for the money. They are created for the everyday needs of users today, offering performance gains and speed for everyday use. In particular, because game developers are constantly pushing the innovation window, the consumer graphics card market is required to keep pace and does so to the benefit of consumers.

  In a nutshell, graphics cards should be fit for purpose. There are plenty of reasons to choose a consumer-grade graphics card; but when it comes to mission critical applications, the benefits to going pro are more highly prized by businesses than raw performance.  Dependability, interoperability and a sustainable long term model – those are the vital elements that make sure the initial investment in pro graphics delivers reliably for years to come.

 

 

Bruno Murzyn is a public relations manager in EMEA at AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied.

 

*Originally Posted by bmurzyn in AMD Business on Mar 11, 2015 4:54:06 AM

The mini-ITX is a popular low power, 17x17 cm motherboard form factor that has been powering a whole range of designs, from complete system solutions for industrial customers to embedded PCs, which include Digital Signage and Thin Client applications, for over a decade now. Fujitsu Technology Solutions, an industry trusted system solution provider and leading vendor of industrial mainboards, once again turned to AMD Embedded G-Series SOC solutions to power their next generation Mini-ITX system design.

 

System solution providers look at a number of different elements when making purchasing decisions around new generation technologies to ensure they are optimizing their system solutions around Total Cost of Ownership (TCO).

 

The pin compatibility in hardware within the AMD Embedded G-Series SOC family across two generations is one of many key contributing factors that enabled customers like Fujitsu to extend their coverage. Our2nd generation product line, codenamed “Steppe Eagle”, gives customers true system scalability in their offerings and takes advantage of not only performance gains but realizes unique feature add-ons like support for configurable TDP (cTDP). In short, the enhanced power management features coupled with two generations of scalable offerings from AMD give hardware and system engineers flexibility in design with respect to thermals while minimizing costs.

 

Another important trend taking shape in Factory Automation is remote management. Remotely accessing enterprise servers and office computers for system health monitoring and maintenance have been a critical function within the IT infrastructure for many years. But these same systems have largely eluded embedded systems on the factory floor due to high costs of implementation and deemed resource intensive… until now. Many of the leading OEMs believe we are in the midst of the 4th industrial revolution or Industry 4.0. Some affectionately call it the “Internet of Things of the Factory Floor”. The end goal for OEMs is to deliver a smart, intelligent and a connected factory floor. Remote management serves as a key enabler to deliver on the promise of a highly intelligent and sustainable factory floor. Factory floor workers can monitor system operations, push software/BIOS updates on the network remotely, among numerous other tasks, driving overall operational efficiency and maintaining factory uptime. AMD, an advocate for proprietary-free open-source solutions, adopted the Desktop and mobile Architecture for System Hardware (DASH) as the ‘gold standard’ for Remote Management. For more information on AMD’s DASH implementation and supporting reference design, check out the webinar from my colleague Cameron Swen and white paper on AMD’s remote management implementation. More details on our own DASH implementation are available via AMD’s Embedded Developer Site. Solution providers like Fujitsu quickly saw the value in an ‘open-source’ royalty free implementation of DASH for remote management for applications like Digital Signage. With mobility on the rise, more customers are looking at differentiated technologies like Remote Management to deliver on their vision for the future.

 

This leads me to the next parameter in the TCO equation: supply. At AMD Embedded Solutions, we continue to listen to our customers’ pain points around their system solution. One recurring theme for Industrial customers is product shelf life and availability. AMD Embedded Solutions has developed a focused Product Longevity Program spanning APUs, SOCs and dGPUs to better serve our broader embedded customer base. The longevity program delivers a supply assurance program eliminating the need to do costly refreshes due to EOL product lines and further builds the relationship with AMD as a trusted partner delivering an optimized TCO solution.

 

Sameer Gupta is segment marketing manager, industrial controls and automation for AMD Embedded Solutions. 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.


*Originally Posted by System Admin in AMD Business on Feb 23, 2015 11:03:29 PM

Attendees of SolidWorks World 2015 in Phoenix, Arizona had the opportunity to witness AMD FirePro™ GPUs flying high alongside key industry leaders to promote our professional graphics features and how they benefit SolidWorks® customers.

 

1.jpg

 

At the week-long conference and exhibition, attendees from around the world discovered how AMD FirePro professional graphics cards provide customers with a professional graphics advantage based on innovation, performance and reliability. AMD FirePro graphics cards provide:

     • Simultaneous render & compute, up to six 4K display support¹, and intelligent power technologies.

     • Certification and optimization for over 100 industry-leading applications including SolidWorks

     • Rock-solid driver support and a three-year warranty with support for the latest APIs and PCIe® 3.0

 

 

Hardware solutions from workstation leaders Boxx, Dell and HP were present in the AMD booth and helped to promote AMD FirePro GPUs in a variety of demonstrations including real-time visualization, collaborating in 3D, professional graphics in the cloud, and support for embedded manufacturing.

 

Attendees were treated to the unique end-user customer participation of a high flying hang glider manufacturer, Wills Wing.

2.jpg

 

 

Producing about 700 hang gliders each year in up to 17 different configurations all hand made by a small team of dedicated enthusiasts, Wills Wing is a leading manufacturer of hang gliders. In addition to certifying each glider to industry standards the company completely assembles and test flies each one as a final quality check before it is dispatched to a customer.

 

 

Steve Pearson, Lead Partner and Chief Designer of Wills Wing said: “The AMD FirePro™ W7000 is phenomenal and now the AMD FirePro W7100™ GPU is giving me a further boost in speed. A couple of years ago such performance would have been completely out of our budget.”

 

Demonstrations provided details about the current AMD FirePro graphics cards with specific focus on advanced SolidWorks workflows and simultaneous engineering, combining complex CAD modeling with sophisticated rendering and simulation (CAM) with ESPRIT from DP Technology. Thoroughly tested, optimized and certified for SolidWorks 2013 and 2014, AMD FirePro professional graphics allow customers to get the most out of the application.

 

“Any 3D engineer or CAM programmer will tell you how graphical-resource hungry CAM software can be while requiring seamless integration with CAD software. Our collaboration with AMD lets us provide our ESPRIT CAM software users and the SolidWorks community with a fully-integrated and performance-optimized CAM solution -- every second counts when you run compute-laded 3D simulations,” said Cédric Simard, Global Marketing & Communication director at DP Technology. “ESPRIT CAM software's adaptive connection to SolidWorks also means any changes in the SolidWorks model are reflected inside ESPRIT and the G-code program. The combination of AMD FirePro graphics with ESPRIT CAM is a great solution for our customers.”

 

 

Attendees also discovered how AMD FirePro professional graphics feature a number of SolidWorks-specific optimizations to ensure designers and engineers get the most out of their workstation. For example, AMD FirePro graphics users can obtain accurate designs with the new GPU-accelerated transparency mode. Order Independent Transparency (OIT) provides a “pixel-accurate” representation of the model and its surrounding geometry and runs much faster than the traditional blended mode because it is accelerated by the AMD FirePro GPU. This creates a more practical transparent 3D viewpoint for designers to continuously work within, helping improve the user’s sense of “design intuition” and aid in better decision-making throughout the product development stages.

 

3.jpg

Learn more about the benefits of using AMD FirePro with SolidWorks

 

 

Antoine Reymond is an industry executive for Professional Graphics at AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied.

OpenCL and the OpenCL logo are trademarks of Apple Inc. used by permission by Khronos.

 

¹4K content required. AMD Eyefinity technology supports up to six DisplayPort™ monitors on an enabled graphics card. Supported display quantity, type and resolution vary by model and board design; confirm specifications with manufacturer before purchase. To enable more than two displays, or multiple displays from a single output, additional hardware such as DisplayPort-ready monitors or DisplayPort 1.2 MST-enabled hubs may be required.  A maximum of two active adapters is recommended for consumer systems. 4K support requires 4K display and content. See www.amd.com/eyefinityfaq for full details.


*Originally Posted by System Admin in AMD Business on Feb 20, 2015 11:03:29 AM

opteron a v2.pngIt’s hard to believe we are already almost two months into a new year – and, even though the ball has dropped, 2015 continues to ring in some exciting times for the AMD software community. We recently announced that popular open source software - OpenSUSE 13.2, Fedora 21, and Xen 4.5-  offer support for our upcoming ARM® server processor, and today I’m excited to announce another exciting innovation in extending the software ecosystem for emerging  server technologies.   The Linux® 3.19 kernel, released on Feb. 9, 2015, includes support for AMD's Heterogeneous System Architecture (HSA) kernel driver (AMDKFD) and the AMD Opteron™ A1100 Series processor (code named “Seattle”), AMD’s server ARM part based on ARM’s Cortex™-A57 core.

 

The AMDKFD driver, which has been publicly under development for the past few months, provides an interface for AMD’s graphical processors for non-graphics (e.g. GPGPU) applications and is the kernel portion to AMD's open-source HSA stack. The user-space side is the new open-source, user-space HSA library and other open-source AMD Radeon™ graphics code. The code is architected to support multiple CPUs each with connected GPUs, with the current implementations focusing on the AMD A-Series “Kaveri” and AMD Opteron™ X-series “Berlin” APUs.

 

HSA is a computing architecture that integrates CPUs, GPUs and other compute devices on the same bus, with shared memory and tasks. HSA, which has been pioneered by AMD, is being developed by the HSA Foundation, which includes ARM and a number of other industry leaders. The goal is to reduce communication latency between these various devices and make them more compatible from a programmer's perspective. The open-source HSA Linux support will continue to be polished over the months ahead.

 

Speaking of ARM, the Linux Kernel 3.19 also adds support for two new ARM64 platforms: the AMD Opteron A1100 Series processor developer kit, and the ARM Juno developer platformAccording to Linus Torvalds, about 20% of 3.19 kernel changes are architectural updates, with these mostly for ARM and ARM64. As previously mentioned, the AMD Opteron A1100 Series processor developer kit is supported by OpenSUSE 13.2, Fedora 21, and Xen 4.5, providing developers with a rich set of operating environments.


It’s certainly an exciting time to be working on HSA and ARM technologies and on the development of innovations that we believe will be highly competitive in the server market.  Keep following us for more updates on our ecosystem progress and new software releases.

 

 

Margaret Lewis is the Director of Software, Server Business Unit at AMD.  Her postings are her 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.


*Originally Posted by klisa in AMD Business on Feb 20, 2015 10:20:00 AM

brystal.boyd

How Do They Do That?

Posted by brystal.boyd Employee May 12, 2015

中文版在英文下方,阅读愉快!

(Chinese translation below English.)

 

There have been several successful television shows dedicated to understanding the many great feats of engineering that have been accomplished. Medical breakthroughs, space exploration, technological marvels; we are fascinated by what we have been able to achieve. At the 2015 International Solid-State Circuits Conference (ISSCC) conference, AMD revealed details on how we accomplished our latest engineering marvel - the upcoming “Carrizo” Accelerated Processing Unit (APU). The semiconductor industry has long relied on axioms of process technology, such as Moore’s Law and Dennard scaling, to drive improvements in device power and performance.  As these laws become more challenging, AMD is responding by implementing a wealth of power management and architecture improvements that in many cases deliver even greater benefit than traditional technology scaling. So, how do we do that?

 

Carrizo Real-estate


The new “Carrizo” microprocessor will include four “Excavator” processor cores and powerful AMD Radeon™ Graphics Core Next (GCN) cores.  With approximately the same area footprint as its predecessor “Kaveri”, “Carrizo” fits 29% more transistors (3.1 billion) onto a die. By utilizing a high-density library design, “Carrizo” achieves a 23% area reduction for the “Excavator” cores while still providing more transistors and more instructions per clock (IPC). The thermal density challenge of the smaller “Excavator” core is mitigated through intelligent floorplan placement and the use of lower leakage transistors. The area reduction for the cores enabled a larger area of the chip to be allocated for graphics, multimedia, and the integration of southbridge and AMD Secure Processor logic onto the APU. The increased footprint for graphics intellectual property (IP) was used to improve the compute performance of “Carrizo,” which is designed to be the world’s first heterogeneous system architecture (HSA) 1.0 compliant part. The multimedia IP has been enhanced with a new high-performance video decoder and double the video compression engines of “Kaveri”. This larger multimedia engine can transcode nine real-time 1080p video streams, an impressive 3.5× improvement over “Kaveri”.

CZ.png

Energy Efficiency and Power Consumption


HSA innovation from AMD saves energy by eliminating connections between discrete GPU and CPU processors, reduces computing cycles by treating the CPU and GPU as peers, and enables the seamless shift of computing workloads to the optimal processing component. HSA allows many workloads to execute more efficiently using GPU compute resources in addition to CPU resources providing better performance at the same energy consumption. Additionally, “Carrizo” moves the GCN cores to a separate conditionally-enabled power supply. This allows the graphics cores to operate at their optimal voltage, which can give a 20% power improvement over “Kaveri” with six GCN cores. “Excavator” supports AMD’s first implementation of adaptive voltage-frequency scaling (AVFS), an improved version of other adaptive voltage approaches. AVFS allows each part to self-calibrate and determine the optimal voltage for current operating frequency and conditions. Timing-margin prediction vs. actual timing margin indicates the ability of AVFS to set the minimum voltage required across the entire voltage range, resulting in up to 30% power savings. The full implementation cost of AVFS is under one percent of the core area. In addition to the area reduction, the “Excavator” core has achieved program goals by reducing power versus the previous “Steamroller” core by 40%!

 

So… How do we do that?


Through a multitude of impressive optimizations, AMD has been able to combine four “Excavator” cores, eight Radeon™ GCN cores, the southbridge, AMD Secure Processor technology for enterprise-class security and a HSA-1.0 design on a single “Carrizo” APU.  The new “Excavator” cores are smaller, more powerful and more energy efficient than the previous generation. The power optimized GCN graphics cores provide impressive performance-per-watt improvements. HSA capabilities enable new, more efficient applications. Multimedia throughput is improved by 3.5x, and hardware support for H.265 decode is included.  All of this is done without a change in process technology, and while holding the die size flat generationally. “Carrizo” is truly a feat of engineering, a great step toward AMD’s 25x20 energy efficiency goal and a testament to the AMD commitment to deliver great products.

 

To dig further into the details, check out the ISSCC 2015 AMD press release and presentation on the ISSCC page of the AMD website.

 

Kevin Lensing is Sr. Director, Client Product Management, Computing and Graphics for AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied.



高能效创新之道

 

 

有好几档不错的电视节目专门展现了人类在工程技术方面取得的许多重大成就。医疗技术突破、太空探索、技术飞跃,我们被人类自己的创举所震撼并为之着迷。即将问世的 AMD “Carrizo” 加速处理器 (APU) 就是一项最新的工程技术成就。AMD 2015 国际固态电路大会 (ISSCC) 上披露了这一产品的一些细节。长久以来,半导体行业依靠工艺技术的规律(例如摩尔定律和登纳德微缩定律)来改善电子器件的功耗和性能。如今这些定律越来越难应验和遵从。AMD 在诸多方面进行了大量的电源管理和架构改进,甚至比利用传统微缩工艺带来的优势更明显,从而很好地应对了这一挑战。那么AMD究竟是如何做到的呢?

 

Carrizo 的基础架构

新款 “Carrizo” 微处理器将包含四个挖掘机”处理器内核多个性能强劲的AMD Radeon™ 下一代图形核心(GCN)。与前代“Kaveri”相比,“Carrizo” 在几乎同样大小的芯片上集成的晶体管数量多出了29% 31亿个)。通过采用高密度库,“Carrizo”使挖掘机内核的面积减少了23%,同时容纳了更多的晶体管并且每时钟周期指令数(IPC)也有所提高。挖掘机内核变小会带来热密度挑战,AMD通过采用智能平面布局和低泄漏晶体管使这一挑战得以解决。内核面积缩小后,APU就会有更大的芯片空间留给集成显卡和多媒体,并用于集成南桥和AMD安全处理器逻辑电路。更大的集显 (专利技术) 面积用于增强 “Carrizo”的计算性能,并使其成为全球首款支持异步系统架构 (HSA) 1.0 的产品。由于采用了新型高性能视频解码器和两倍于“Kaveri”的视频压缩引擎,多媒体 (专利技术) 性能也得到了增强。更强大的多媒体引擎可以转码9个实时1080p视频流,比 “Kaveri”提升了惊人的3.5倍。

 

能效与功耗

HSA AMD 的一大创新。它通过消除独显和CPU之间的连接节省了能源,通过把CPUGPU放在对等的位置减少了计算周期,而且它能够把计算负载妥善地转移到最优的处理部件上。HSA能够利用GPU来更高效地执行许多负载,而不单纯依赖CPU资源,在相同能耗的情况下带来更好的性能。此外,“Carrizo”GCN内核采用独立的、按条件启用的电源,因此图形核心可以工作在最佳电压之下,其功耗与具有6GCN核心的 “Kaveri”相比改进了20%挖掘机支持AMD 率先应用的自适应电压-频率调节 (AVFS) 。相比其他自适应电压方案,AVFS经过了改进。它让每个部件都能够根据当前的工作频率和状态进行自我调节并确定最佳工作电压。通过比较预估时钟余裕和实际时钟余裕可以看出AVFS在整个电压范围内确定所需最低电压的能力。这一技术可节能高达30%。而完整实施AVFS的仅占用不到1%的内核面积。“挖掘机内核不仅减小了面积,而且还实现了计划目标:功耗比压路机内核降低40%

 

 

那么,效果如何呢?

经过大量令人赞叹的优化之后,AMD已经能够将四个挖掘机内核和八个Radeon™ GCN内核、南桥、面向企业级安全应用的AMD安全处理器技术、以及HSA-1.0设计,集成到一个“Carrizo” APU中。新的挖掘机内核比上一代更小、更强大并且更节能。功率经过优化的GCN图形核心的每瓦性能提升惊人。HSA特性带来新的、更加高效的应用。多媒体处理能力提升了3.5倍,并且支持 H.265 硬解码。如此多的革新和进步并未通过改变工艺便已经实现,而且芯片尺寸依旧纤薄小巧。“Carrizo” 真的是工程技术的精妙杰作,也是AMD朝着 25x20 能效目标迈出的一大步,更是 AMD 履行创造伟大产品承诺的有力证明。

 

 

如欲了解更多详情,请参见ISSCC 2015 AMD新闻稿和AMD官方网站的 ISSCC 讲稿。

 

Kevin Lensing AMD 图形与计算事业部的客户端产品管理高级总监。他的博文仅代表他个人观点,不代表AMD的立场、战略或观点。指向第三方网站的链接仅供方便您阅读之目的,除非特别注明,AMD不对指向的网站上的内容负责,也不意味着AMD对其内容持赞同态度。


*Originally Posted by System Admin in AMD Business on Feb 23, 2015 6:53:29 PM

Guest Post By Tommy Swigart, Global Product Manager COMExpress, GE Intelligent Platforms

 

At GE, we’ve already invested over $1 billion in making the Industrial Internet a reality. It’s here and now – and it’s transforming businesses around the world. Of course, its continued growth won’t be as the result of a single company’s effort and investment; rather, it will be because companies have worked together. No single company has every piece that’s needed.

 

Here’s a case in point. At GE, we’ve identified COM Express technology as a key building block within an infrastructure that brings together advanced computing, analytics, low cost sensing and new levels of connectivity. COM Express has three highly desirable attributes. It’s modular and easily upgradable; it’s an open standard; and, in the right hands, it’s capable of withstanding the rigors of deployment in the harshest industrial – and military – environments. That’s why COM Express is at the heart of GE’s next generation industrial PCs and automation controllers – machines that will deliver the underlying intelligence that will help drive the Industrial Internet.

 

One of the other beauties of COM Express technology is that it’s processor-agnostic; it can be a carrier for virtually any processor, making it hugely flexible and adaptable.

 

So: a customer came to GE, looking for a very challenging solution. He needed something with all the attributes of COM Express. He needed, however, to pair the COM Express carrier with a processor with very low power consumption – but with substantially more capability than is typically available from processors designed for such applications. The customer also needed the most compact package possible.

 

Not only that: rugged reliability, with maximum possible MTBF (mean time between failure), was also crucial to his application.

 

At GE, we’re fortunate to have a very close working relationship with AMD, and we were aware of AMD’s plans to launch their G-Series System-on-Chip (SoC). It had the attributes that we – and the customer – were looking for. Not only did it deliver the high performance/low power consumption needed by the application, but it also included much of the functionality that would otherwise need to be implemented on the underlying COM Express carrier. The impact of this, of course, is that the lower component count increases MTBF.

 

And so it was that the mCOM10-L1500 was born – a result of close cooperation between the teams at AMD and GE. AMD’s G-Series technology provided a combination of functionality and performance characteristics that enabled the development of a product uniquely suited to customer demands.

 

1-Optimized-GE-enables-the-Industrial-Internet-with-AMD-Embedded-G-Series-SOC_Frame89.jpg

The Mini COM Express module mCOM10-L1500 offers high performance and ultimate durability needed for applications that operate in harsh environments.


The inherent reliability of the AMD G-Series SoC is complemented by GE’s fully rugged design. Onboard components are specifically selected for their reliability in demanding conditions, and are soldered to the board for maximum resistance to shock and vibration, while extended mechanical construction protects the module. The mCOM10-L1500 is also designed to accept conformal coating for even greater resistance to moisture, dust, chemicals and extremes of temperature.

 

Like all COM Express-based designs, the mCOM10-L1500 delivers lower lifetime cost of ownership, because upgrades to the processor – in response to changing application demands, or to leverage new generations of price/performance – are straightforward and minimize cost. It is also the case that, as the underlying carrier card that provides the required interfaces to the system does not need to be replaced at the time of upgrade, testing and requalification time, effort and expense are minimized.

 

A single, new product developed by the teams at GE and AMD will not, in itself, ensure the continuing growth of the Industrial Internet. It will take hundreds – thousands – tens of thousands – of additional new building blocks – but products like the mCOM10-L1500 will make a vital contribution to our growing ability to connect people, data and machines.

 

 

Tommy Swigart can be contacted at Thomas.Swigart@ge.com.

 

Tommy Swigart is Global Product Manager COMExpress at GE Intelligent Platforms. 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.


*Originally Posted by System Admin in AMD Business on Feb 23, 2015 10:03:29 AM

Guest Post By Jos Schellevis, Chief Technical Officer, Deciso

 

The ever-expanding connectivity of applications brings new security threats that require a different defense than traditional firewalling.

 

The focus of hackers and cybercriminals has shifted from network intrusion attempts to exploiting weaknesses in applications.

 

These application-layer attacks also increasingly make use of encryption to dodge network security defenses.

 

A recent Gartner report titled “Security Leaders Must Address Threats From Rising SSL Traffic” by Jeremy D’Hoinne and Adam Hils, states: “Gartner believes that, in 2017, more than half of the network attacks targeting enterprises will use encrypted traffic to bypass controls, up from less than 5% today.[i]

 

These threats demand more powerful hardware to decrypt and detect intrusion attempts at wire speed. Next generation firewalls not only have the capabilities to detect application-layer attacks but also have sufficient power to accomplish this task at gigabit connection speeds.

 

Any hardware design to accomplish this task thus requires a multicore CPU and fast Ethernet connections without any bottlenecks. Many existing designs have a separate CPU and chipset connected through a marginal interface, not capable of leveraging its full performance over the external – mostly PCIe® – interfacing.

 

To make things worse, some of the network designs we have come across in the past have integrated PCI bridges creating even more bottlenecks. While this may not have been an issue in 10/100Mb solutions, today’s networks increasingly operate at gigabit or even higher wire speeds.

 

The highly integrated AMD G-Series SOC at the heart of Deciso’s Netboard A10 design delivers the required performance, doesn’t suffer from bottlenecks and has low power requirements. The embedded low power design also eliminates high cooling requirements that current high performing server-like designs demand. And, the integrated AESNI engine makes encryption and decryption of VPN traffic much less CPU intensive. While new features of the next generation firewall may consume quite a bit of the available CPU cycles to inspect application-layer traffic, other demanding features such as VPN won’t suffer.

 

Deciso believes that the combination of a great design and AMD G-Series SOC makes the Netboard A10 a solid bases for next generation firewalls.

 

Jos Schellevis is Chief Technology Officer at Deciso B.V, a Dutch security equipment manufacturer. He graduated in workflow management at Rotterdam University of Applied Technology and has over 15 years experience in networking and telecommunications. 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.

 


[i] https://www.venafi.com/collateral/wp/gartner-responding-to-new-ssl-cybersecurity-threats

 

*Originally Posted by System Admin in AMD Business on Feb 11, 2015 2:10:40 PM

opteron a v2.pngWhenever I talk about the AMD Opteron™ A1100 Series processor (codenamed “Seattle”) one of the first questions I get asked is “how is the software ecosystem for ARM servers maturing?” It is an established fact that the success of ARM® servers requires the right combination of software to support today’s demanding data center workloads.  It’s also a fact that finding that perfect combination doesn’t just happen overnight – and as you likely know, we’ve dedicated significant time, energy and resources into building a stable ecosystem for ARM servers….and the fun has only just begun for 2015.

 

I am happy to report that the recent releases of Fedora 21 and Xen 4.5 will support the AMD Opteron™ A1100 Series processor! These are two significant milestones toward achieving our goal of a robust software ecosystem to drive the adoption of ARM server technology. Fedora 21 and Xen 4.5 join OpenSUSE 13.2 in providing developers using AMD Opteron™ A1100 Series 64-bit ARM Developer kits a growing base of operating systems and hypervisors to choose from.

 

In December, the Fedora ARM Team announced the release of Fedora 21 for AArch64, a game-changing technology ready to run in the emerging world of ARM servers. Fedora 21 for AArch64 delivers the software infrastructure needed to run basic server applications stacks, such as storage and file serving (Ceph, GlusterFS, XFS), Web server (Apache, Tomcat), database server (MySQL), and developer tools (OpenJDK, Phython, Perl). It also includes KVM for developers who want to take an early look at virtualization on ARM server platforms. Fedora is developed by the Red Hat-sponsored Fedora community, providing users with access to their latest free and open source software.

 

In January, the Xen Project announced the release for Xen 4.5. Spanning across x86 and ARM architectures and several server platforms, Xen 4.5 features a hypervisor code base with increased usability, simplicity and innovation. The Xen Project provides an overview of its ARM Hypervisor-Specific updates including UEFI booting, IOMMU, and Super Pages support.  While not all features made this release, Xen 4.5 provides the basics needed to begin creating virtualization environments on ARM servers. The Xen Project is a collaborative project of the Linux Foundation, a nonprofit consortium dedicated to fostering the growth of Linux® and collaborative software development.

 

2015 promises to be a big year for the ARM64 software ecosystem, so be sure to stay tuned for further developments. Industry leaders in Linux on ARM - including AMD, the Linux Foundation, and Red Hat - will be gathering in February in Hong Kong for Linaro Connect  to showcase the latest in software developments and optimizing ARM technology.

 

Margaret Lewis is the Director of Software, Server Business Unit at AMD.  Her postings are her 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.

 

*Originally Posted by klisa in AMD Business on Jan 29, 2015 5:33:00 PM

For many of us here at AMD, the start of a new year always coincides with the biggest technology tradeshow in the world: the International Consumer Electronics Show.

 

For AMD Embedded Solutions, CES 2015 was particularly exciting as our technology partners at QNAP, a leading provider of network attached storage (NAS) solutions, announced a new family of products powered by AMD Embedded Solutions

 

AMD-Facebook1.jpg

 

QNAP’s new TVS-863+ and the TVS-x63 NAS systems incorporate the AMD Embedded G-Series system-on-chip (SoC). This platform, designed for small to medium-sized business and small office/home office customers, is a leader in its class, powered by an AMD Embedded G-Series 2.4 GHz quad-core SoC with 10 GbE capability, legendary AMD Radeon graphics, and AMD-V™ virtualization support. Loaded with features, the new NAS systems can help lower the operational costs and increase productivity when used in 24/7 business applications. You can learn more about this system and the AMD Embedded technology powering it here.


Along with the new NAS systems, we also showed off Gizmosphere’s Gizmo 2 DIY platform in a home media center environment. For this demonstration we had Gizmo 2 streaming video from the QNAP TVS-863+ via UPnP (Universal Plug and Play) and using the free/open-sourced Kodi (formally known as XBMC) media player software. Using Gizmo 2 as a media center is just one of seemingly countless ways to leverage the capabilities of this new DIY platform.

 

Ideal for the embedded programmers and advanced DIYers, Gizmo 2 is an open source development board, powered by the AMD Embedded G-Series SoC, offering outstanding compute and graphics performance on a single platform for a wide range of Linux® and Windows® based development projects.

 

Gizmo 2 will be available for purchase worldwide from Gizmosphere in February for $199 (USD) – check Gizmosphere.org for updates.

 

Check out this video to see Gizmo 2 in action:



After an enthusiastic showing at CES, we are energized for 2015 and look forward to sharing more highlights as we help bring more products powered by AMD Embedded Solutions to market.

 

Travis Williams is the product marketing manager for AMD Embedded Solutions. 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.


*Originally Posted by traviswilliams in AMD Business on Jan 14, 2015 5:04:38 PM

Medical imaging professionals are currently challenged with ‘information overload’ when it comes to critical patient image and scan results. Too often, they must view different patient results on multiple monitors that sometimes exist in different locations. Because medical professionals are responsible for reading patients’ X-rays, MRI, CT or mammography scans, they truly understand and appreciate the importance of accurate graphics and display quality.

 

With thirty years’ experience in the medical imaging industry, Barco relies on high quality graphics cards (display controllers) for their medical displays. Our collaborator, for more than a decade, is leading innovation and smart technology breakthroughs to deliver incredible image quality for radiology, mammography, surgery, dentistry, and modality imaging. Their healthcare division looks to AMD professional graphics products to power its diagnostic display lines. Barco’s latest innovation, the Coronis Uniti™, is designed with radiologists in mind by offering unparalleled image quality, inventive productivity features and ergonomic design.


photo1.png

Image courtesy of Barco

 

According to Barco, at 12 megapixels, Coronis Uniti - powered exclusively by Barco MXRT, built on AMD FirePro™ W7000 graphics cards - is the highest resolution display available in the medical imaging market today. Through DisplayPort 1.2 support, AMD FirePro W7000 graphics help Coronis Uniti to achieve its massive maximum resolution of 4200x2800. That’s 12 million pixel points in a 33-inch display, enabling sharp grayscale and bold, brilliant colors for viewing 2D/3D static or dynamic images in in-depth detail. The display is also equipped with Barco’s industry-changing SteadyColor™ calibration technology, and Color Per Pixel Uniformity™ to ensure consistent colors and grayscales, along with Ambient Light Compensation™ to ensure that image quality is flawless, in a variety lighting environments.

 

In designing Uniti, Barco strived to deliver exceptional image quality while easing radiologists’ physical discomfort when reading images. The display optimizes the reading experience by mirroring our natural field of vision. Its carefully designed format minimizes the need for head and eye movements, while also creating side-by-side comparisons of multiple images. To reduce eye fatigue, its SoftGlow™ wall light adds ambient light to the reading room, while the SoftGlow™ task light shines a light on papers and film folders, while SpotView™ focuses light and allows radiologists to view an area more closely.


photo2b.jpg

Image courtesy of Barco


Image quality and consistency are paramount to achieving a confident diagnosis, and ultimately, a successful course of treatment for patients. Barco technologies that go beyond pixels and lumens enhance even the subtlest details to aid in precise detection at the earliest possible stage. AMD and Barco have worked together for nearly 12 years to enable cutting edge features like those found in Coronis Uniti. Barco expands on the rock-solid AMD FirePro unified driver and its features to develop and incorporate innovations unique to its diagnostic displays and software, resulting in the superior visual experience for the medical imaging community.

 

To learn more about the Coronis Uniti powered by AMD FirePro W7000 graphics, check out this video.

 

Theresa Chavez is a product marketing manager, professional graphics at AMD. Her postings are her own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied.


*Originally Posted by System Admin in AMD Business on Nov 12, 2014 11:21:42 AM

It wasn’t too long ago that GPUs only focused on graphics and single-precision performance and either didn’t support double-precision operations, or only supported them at dramatically reduced performance levels. With the AMD FirePro™ S9150 server GPU, AMD now delivers exceptional compute performance with its newest and most powerful server GPU ever built. Based on our latest Graphics Core Next architecture, the AMD FirePro S9150 delivers maximum double-precision performance, driving circles around the competition, offering up to 77% more double-precision performance than the Tesla K40.1

 

Our performance superiority has recently been demonstrated using DGEMM, with performance measurements taken from a single AMD FirePro S9150 GPU.  DGEMM, or Double-precision GEneral Matrix-Matrix, measures the floating point execution rate for double precision real matrix-matrix multiplication. DGEMM computations are part of the BLAS (Basic Linear Algebra Subprograms), a specified set of low-level subroutines that perform common linear algebra operations such as copying, vector scaling, vector dot products, linear combinations, and matrix multiplication.


dgemm diagram.png

 


There are many real-world applications that take advantage of double-precision matrix operations.  These include computational fluid dynamics, finite element analysis and structural modelling, and molecular dynamics.


With our AMD OpenCL™ BLAS implementation, we are able to achieve 2 TFLOPS of sustained DGEMM performance with the AMD FirePro S9150.  This is a first for a single GPU solution, and with the Tesla K40 only achieving a theoretical 1.43 TFLOPS of peak double-precision, we are able to show the world that AMD can win by a wide margin when comparing actual measured results versus Nvidia’s theoretical performance.1 With industry-leading performance/watt2, industry-leading memory configuration3, and support for the latest OpenCL™ standards4, the AMD FirePro S9150 Server GPU is clearly unmatched when it comes to compute performance. 

 

AMD is serious about HPC and we want to show you that we are able to lead in this space. We are focused, we are committed, and most importantly, we are here to stay.

 

The AMD FirePro S9150 server GPU is available for purchase today. Please visit http://www.amd.com/en-us/products/graphics/workstation/firepro-remote-graphics/s9150 for more details.

 

JC Baratault is a senior business development manager, Global GPU Computing for professional graphics at AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied.


OpenCL and the OpenCL logo are trademarks of Apple Inc. used by permission by Khronos. PCI Express is a registered trademark of PCI-SIG.

 

1.AMD FirePro™ S9150 delivers up to 2.53 TFLOPS peak double precision floating point performance, and Nvidia’s highest server GPU in the market as of June 2014 is the Tesla K40 with up to 1.43 TFLOPS peak double precision. Visit http://www.nvidia.com/object/tesla-servers.html for Nvidia product specs. FP-95

 

2. AMD FirePro™ S9150 max power is 235W and delivers up to 2.53 TFLOPS peak double and up to 5.07 peak single precision floating point performance. Nvidia’s highest performing server cards in the market as of June 2014 are the Tesla K40, max power of 235W, with up to 1.43 TFLOPS peak double and up to 4.29 peak single, and the K10, max power 225W, with up to 4.58 TFLOPS peak single and 190 GFLOPS peak double precision. Visit http://www.nvidia.com/object/tesla-servers.html for Nvidia product specs. FP-97

 

3. AMD FirePro™ S9150 features 16GB GDDR5 memory, and Nvidia’s highest performance server GPU in the market as of June 2014 is the Tesla K40 with 12GB GDDR5 memory. Visit http://www.nvidia.com/object/tesla-servers.html for Nvidia product specs. FP-98

 

4. OpenCL™ 1.2 conformance expected for S9150 and S9050. AMD plans to release OpenCL™ 2.0 drivers for enabled AMD FirePro S9150 server GPUs in Q4 2014; conformance testing is planned at that time. Previous generation AMD FirePro products may not support OpenCL™ 2.0.


*Originally Posted by System Admin in AMD Business on Nov 11, 2014 12:01:42 PM

The PC is dying… or so goes the opinion of some. It is true that there has been some reduction in overall PC unit volumes in recent years, although many current forecasts indicate that despite weak consumer PC volumes, commercial client PC units are growing. AMD believes we have a tremendous opportunity to grow share in the commercial client space based on our products and current market trends and have doubled our number of commercial design wins this year compared to last.

 

52914B_AMD_Pro_E_RGB.pngOne of AMD’s goals is to derive approximately 50% of our revenue in high-growth adjacent markets by the end of 2015. That growth is dependent upon our core IP in CPU and GPU compute products and we continue to drive innovation in those areas. Our Compute and Graphics unit is focused on the areas where our IP gives us differentiation and we can be competitive. It’s to that end that AMD introduced AMD PRO Series products in June 2014.

 

New devices entering IT infrastructures are based on various operating systems and processor types. This has introduced new problems for IT managers as what was once a (relatively) simple infrastructure in terms of security and manageability became increasingly complex. Bring your own device (BYOD) became the bane of many IT managers lives. As a result, older, proprietary security and manageability solutions need to be replaced by industry standard, cross-platform solutions.

 

This shift from closed/proprietary standards to open standards is where AMD is focused. AMD is a member of the Distributed Management Task Force (DMTF) which is “an industry standards organization working to simplify the manageability of network-accessible technologies through open and collaborative efforts by leading technology companies.” DMTF is the organization behind the Desktop and mobile Architecture for System Hardware (DASH) Standard, “a suite of specifications that takes full advantage of DMTF’s Web Services for Management (WS-Management) specification – delivering standards-based Web services management for desktop and mobile client systems.” AMD has also recently introduced AMD Secure Technology utilizing the Platform Security Processor (PSP) to enable ARM® TrustZone providing a secure platform for service and content. AMD is enabling evolving specifications like DASH and TrustZone that today’s IT managers need to maintain secure and manageable infrastructures.

 

We believe AMD is well positioned for growth in commercial client PCs and are seeing some positive indicators. DASH enabled AMD-powered systems have been growing in numbers. With one of the most ubiquitous management consoles, Microsoft® System Center 2012 Configuration Manager (SCCM) supporting DASH manageability, the AMD PRO solution is a great fit for existing IT infrastructures moving away from proprietary “solutions.” The very first AMD PRO-powered touch enabled and thin & light commercial client systems have been launched. In the Q3 2014 earnings, AMD announced that we have met our goal to double our commercial client design wins from last year.1

 

AMD is known as a graphics leader and this is helping meet the demands of today’s commercial client system user. Gone are the days when data could only be presented in a spreadsheet. In today’s fast-paced world, information reporting can be done in easily digestible, highly visual formats as described in a recent Jon Peddie Research article. Video is increasingly being used to convey thoughts and ideas. Desktop phones are being replaced with PC-based video conferencing applications. Most system users are not concerned with how things happen, only that they do happen and that the experience is great. The methods of evaluating systems used for purchase tenders has been changing to utilize more modern benchmarks that address total system capabilities, like Futuremark’s PCMark® 8 v2. IT departments are increasingly working with more strictly controlled or limited budgets making the needs to use these new benchmarks more and more important to show that they are considering everything relevant to performance in their purchase decisions. The IT managers’ internal customers are expecting devices that look and feel like their personal devices, thin, light and responsive. Feedback from customers on the AMD PRO-based HP Elite series systems and how the systems meet their needs today has been quite positive.

 

You can find more information on AMD’s renewed focus on delivering the complete solutions required by IT managers in our newly revamped AMD PRO website. It may be true that PC sales are not going to experience the traditional growth as has been seen through history. But industry analyst forecasts of PC sales appear to be somewhat stabilizing… and AMD is poised, with the AMD PRO solution, to deliver the solutions that IT managers need.

 

Ryan White is a Product Manager – Commercial Client 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.

 

1 See http://phx.corporate-ir.net/External.File?item=UGFyZW50SUQ9MjU1MDY0fENoaWxkSUQ9LTF8VHlwZT0z&t=1

 

*Originally Posted by System Admin in AMD Business on Nov 3, 2014 2:01:42 PM