While the memo specifically addresses challenges in the mobile sector, it delivers a powerful message on the importance of ecosystems that can easily translate to our industry. The narrative suggests that in order to remain competitive, companies need to compete collectively by maintaining a sustainable ecosystem of partners. With respect to the embedded market, we at AMD believe that the key to building a sustainable business is to continue to invest in technology and programs that allow us to anticipate our customers’ needs and that will help bring their products to market faster and more efficiently.

Embedded Systems, Ecosystem, G-Series, Fusion Partner Program, Coreboot, Reference Design Kit, x86We’ve already begun to make significant headway toward strengthening our commitment to our ecosystem by announcing the AMD Embedded G-Series Platform of AMD Fusion products that will better serve the needs of our customers. We continue to build out this product line, and this week we updated the platform to include “Bobcat” CPU options for headless and traditional embedded systems that brings the benefits of our newest core and the x86 instruction set to even broader range of customers.  Meanwhile, our new Reference Design Kit (RDK) ensures innovative system design support and releasing the source code for coreboot supporting open source software development continues our proud commitment to choice in the ecosystem. After the products have been developed, we now provide additional sales and marketing solutions through our expanded AMD Fusion Partner Program. 

The approach we’ve taken to nurture multiple aspects of our ecosystem is evidence that we are maturing our embedded business. We understand the pressures that our partners are under, and we’re continuing to stay ahead of the game in order to deliver on our commitments and continue fostering innovation and success.

Buddy Broeker is the director of AMD Embedded Solutions. 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. 

Let’s take a brief tour of the 4P market before we show you how we are really shaking things up.

The 4P server market has traditionally been the place where large databases, consolidated virtualized infrastructures, business applications and large e-commerce engines would live.  Customers paid more, a lot more, in proportional terms, to get their hands on these platforms.  One could justify the expense based on the high performance and large memory footprint.

Back in 1996, the 4P market was 13.1% of the total x86 market and at the end of 2009 it stood at only 4.2% (IDC Server Tracker, 4Q 2009). But without some serious change in the market, all of my experience tells me that on the current trajectory we would exit 2010 with a “3” at the front of the 4P market mix percentage.

This is why AMD called for the death of the 4P tax.

You see, 4P capable processors were always priced a lot higher than their 2P counterparts, even if the performance (socket to socket) was the same, or even, at times, lower. When we first introduced the AMD Opteron™ processor back in 2003, we shrunk that 4P tax, but it was still there.  Entering the market with a smaller premium allowed AMD to gain share in the 4P market. But there was still more work to do.

With the latest generation of processors, you’ll find that our top bin 2P capable processor is priced at $1,386USD*.  Our top bin 4P capable processor also is priced at $1,386USD.  This is no coincidence, because they are the same processor (6100 series), with the same chipsets, drivers and supporting architecture.

Our competitor has chosen to take a different path. With their previous generation (2P Xeon 5500 /4P Xeon 7400) their top bin comparison was $1,600/$2,729USD (as of 5.26.2010).  This represents a 70% premium (or as we sometimes style it, a “tax”) to move from the competition’s 2P capable system to their  4P capable system.  With their latest generation, that 4P tax has jumped up dramatically.  Today’s Xeon 5600/7500 comparison is now $1,663/$3,692USD (as of 5.26.2010). This represents a 122% “tax” that customers have to pay to go to 4P.  And we wonder why the market is shrinking?

Where has this 4P market gone?  The 4P market has grown in units, but not at the rate of the overall server market. As the x86 server market has grown over the past few years, the 2P platforms saw all of the growth because customers found the reliability and scalability meeting their needs.  The price/performance couldn’t be ignored.

To save that market, something had to fundamentally change, which is why we introduced the concept of the “Value 4P”, a platform designed to bring the performance of 4P with the value more associated with a 2P system.

To give you an idea of how this plays out in real products, I took a look at the new Dell PowerEdge™ R815, a 4P AMD-based platform.  This 2U 4P server is a great solution for not only 4P workloads, but also many high end 2P workloads. This raises the question – if you could get 4P performance at 2P prices, why wouldn’t you do that?

I configured a few systems to prove my point

We held the configuration constant to help isolate the impact of processor pricing on the overall system price.

A 4P priced UNDER a 2P?  Up to 4X the cores for a lower price (12 cores x 4 processors = 48 total cores vs. 6 cores x 2 processors = 12 total cores)?  Suddenly there are a lot of places to think about 4P technology where previously 2P was the only realistic option.  With 32 total DIMMs sockets, the Dell PowerEdge R815 can support a much larger memory footprint than the typical 2P systems, which commonly feature just 18 total DIMM sockets.

Where does this platform make the most sense?

• HPC clusters. If you are not saturating the fabric interconnect, wouldn’t you rather have a single 4P node instead of two 2P nodes?  This would allow you to cut the interconnect card cost and your switch and cabling costs because you’ll have half as many nodes. Plus, you could see lower overall power and management costs by using half the nodes.
• Databases. Microsoft SQL server licenses by the server, up to 4P, so license cost is the same for either a 2P or 4P system. But with 4P, you get the benefit of additional cores to help speed up your multiple simultaneous queries.
• Consolidation. If you are consolidating workloads, wouldn’t 48 cores in only 2U of rack space meet your needs for scalability and performance in a condensed form factor?
• Virtualization. More cores and more available memory sockets make this a great virtualization platform.  In addition, having the potential for 2 SD cards to run the hypervisor can make this a great virtualization engine for your business.

As you think about your next 2P server, maybe it is time to consider the new Dell PowerEdge R815, which would mean your next 2P server could actually be a 4P system.  And if you can get 48 cores for less than cost of 12, the options for deployment expand vastly.

With choices like this new “Value 4P” platform and the removal of the “4P tax”, we might actually see that 4P market trend turn around. And who can argue with getting more for your money?

For more information, download our 4P white paper.

Configuration 1: Dell PowerEdge R815, 4x AMD Opteron™ 6168, 1.9GHz, 12C, 32GB Memory (16x2GB), 1333MHz, Dual Ranked UDIMMs, 146GB 10K RPM Serial-Attach SCSI 6Gbps 2.5in Hotplug Hard Drive, No Operating System, Sliding Ready Rails With Cable Management Arm, 3Yr Basic Hardware Warranty Repair: 5×10 HW-Only, 5×10 NBD Onsite, 1100 Watt Non-Redundant Power Supply, $8,065.00 as of 5/20/2010 http://www.dell.com

Configuration 2: Dell PowerEdge R815, 4x AMD Opteron™ 6136, 2.4GHz, 8C, 32GB Memory (16x2GB), 1333MHz, Dual Ranked UDIMMs, 146GB 10K RPM Serial-Attach SCSI 6Gbps 2.5in Hotplug Hard Drive, No Operating System, Sliding Ready Rails With Cable Management Arm, 3Yr Basic Hardware Warranty Repair: 5×10 HW-Only, 5×10 NBD Onsite, 1100 Watt Non-Redundant Power Supply, $8,065.00 as of 5/20/2010 http://www.dell.com

Configuration 3: Dell PowerEdge R710, 2x Intel  Xeon  X5680, 3.33Ghz 6-core,  32GB Memory (8x4GB), 1066MHz Dual Ranked RDIMMs, 146GB 10K RPM Serial-Attach SCSI 6Gbps 2.5in Hotplug Hard Drive, No Operating System, Sliding Ready Rails With Cable Management Arm, 3Yr Basic Hardware Warranty Repair: 5×10 HW-Only, 5×10 NBD Onsite, High Output Power Supply, Non-Redundant, 870W, $8,244.00 as of 5/20/2010 http://www.dell.com

*Based on AMD 1Ku pricing

John Fruehe is the Director of Product Marketing for Server/Workstation products 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.

If you closely examine the markets where industry-standard x86 AMD Embedded Solutions are winning major designs, you’ll see these are high-performance systems.   Our friends at Quixant are looking for a combination of factors in selecting their gaming platform, including multi-core CPUs for demanding apps, the ability to drive multiple displays, the graphics to ensure those displays look fantastic and of course, they need a strongly competitive price.

We have other customers, iBASE for example, who rely on our ability to provide consistency in our top-to-bottom platforms.  They can depend on performance improvements with additional cores, new features and consistent power envelopes to help extend the life of their embedded designs.

Throw in the fact that these companies and many more are increasingly building their products around high definition graphics and multiple displays and it is pretty clear that the complete AMD embedded platform – with enterprise-class CPUs, chipsets and graphics – fits the bill.

So while we see various competitors in this huge marketplace trying to push into new markets, we have been working closely with our customers to deliver on their needs.   We’re doing that today with our two new complete platforms and as we look to the expected advent of AMD Fusion™ technology products in 2011, our position for helping drive down power, area and costs while improving the design and user experience will just get stronger.

Buddy Broeker is the director of AMD Embedded Solutions. 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.

They started their journey weeks ago, in the German state of Saxony, where the GLOBALFOUNDRIES fab began the process of building these marvels of modern silicon technology. Precision German engineering (I am biased after all) is in the heart of making one of the greatest innovations in processor technology.

The wafers begin life as non-descript silicon, but over the weeks of the process, the 8 and 12-core “Magny-Cours” begin to take shape.  After the wafers are finished, they head on to Penang, Malaysia for packaging before the final stop in Singapore.  There, the test, marking and sort happens, with processors ending up in those trays that you’ve seen so often.

Production began last month and our OEM partners have been receiving production parts this month.  We have had a few select end customer opportunities that have been fulfilled, but it is nothing we can talk about publicly.

As a matter of fact, we were not planning to talk about any of this just yet, but earlier this week, someone tried to offer products that they claimed were “Magny-Cours” processors for sale on the web.  Obviously the message about 12-core goodness with incredible value is making it out into the market, so much so that somebody wanted to jump on the bandwagon.

I wasn’t really expecting to write this blog just yet, I was hoping to spring that news sometime in the near future, but the internet’s “series of tubes” have been lit up and stole my thunder.

So, there it is, we’re building them now, and we expect that you’ll be seeing a launch before the end of the quarter.  With 8 or 12 cores, 4 memory channels and lots of other great technology coming, it’s worth the wait.  WELL worth the wait.

John Fruehe is the Director of Product Marketing for Server/Workstation products 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.

The list is a relative “who’s who” of the microprocessor world.  Every processor clearly had a huge impact on the market, and it is an honor to be included in the list. Most importantly, we are the only processor in the list that was released after 1993. AND we are the only one that is still in the market today in any reasonable volume (somehow, somewhere, somebody is probably utilizing the others because they were such strong designs).

Back in 2002, a year before the launch of the AMD Opteron processor AMD and Microsoft announced a cooperative effort to develop 64-bit version of windows for AMD 64-bit processors. We were the first mainstream processor to enable 64-bit, and our shared vision was that 64-bit computing would drive the data center of the future.  I guess you could say that we helped influence the move to 64-bit.

In 2003, there was no such thing as a “low powered server processor.” Enterprise-class processors were big, hot and power hungry.  Now look at where we are today.  The new AMD Opteron EE processor delivers six cores in a single processor, with a 40W ACP. Power is one of the biggest concerns in the data center today, and we influenced the market with the introduction of power efficiency for servers back before people realized how important power would be.

What about price? If you can remember the pre-AMD processor Opteron days (or the “dark ages” as I like to call them), there was little competition in x86 server processors.  Price/performance was a very limited concept, your price/performance came at the platform level, not really at the processor level. Prices were significantly higher than they are today.  AMD influence? Check.

Remember the world of single core computing? It may be hazy for most of you because for the past 4 years, almost every server processor sold was probably a multi-core.  And who had the first multi-core X86 processor?  Yep, the influencer.

So what does the server processor of 2009 look like? Well, let’s take a look at the profile:

• Multi-core
• 64-bit x86
• Integrated memory controller
• Low power options
• High performance per watt
• High price performance
• Hardware-based virtualization capabilities

When you put the list together, it is clear that AMD has had a significant impact on the server processor market.  Regardless of which processors you are deploying, what you enjoy today in your X86 data centers are either AMD Opteron processors, or processors that have been heavily influenced by our technology decisions. Some would say that “imitation is the sincerest form of flattery.”  I like to call it influence.

John Fruehe is the Director of Business Development for Server/Workstation products 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.

It’s hard to believe that only a few short years ago, hardware assisted virtualization was not generally available on the x86 platform. The intervening years have brought an enormous amount of innovation that translates directly into economic value to customers. And we’re not done yet.

While some may be looking for the next big problem to apply virtualization, many IT organizations want to keep on doing what they’re doing, only better. What is holding them back? From a workload perspective it is usually support, performance and risk. Support and risk can be lumped together as problems that require more than technology to solve. Workload performance has technology written all over it.

It used to be a truism that you could only virtualize the smallest, least mission critical workloads. But we read every day about organizations that have success virtualizing ever larger workloads. However, there are still some workloads that don’t fit comfortably in virtual hardware. If we want to virtualize everything, we need to reduce virtualization overhead.

Overheads are an inevitable part of virtualization; each layer of abstraction introduces another step to the dance. The trick is to identify which of the extra steps takes the most time, or is performed the most frequently. And then to find ways to remediate these overheads. One step along this path is optimizing SMP (i.e., more than 1 CPU) workloads. Once you move to a second CPU, there are all sorts of interesting issues introduced. As we tested larger workloads, we noticed that virtual CPUs were sometimes scheduled when they shouldn’t be. This caused extra processing without any benefit.

After analyzing the problem, we made a change that resulted in substantial performance improvements. But we worried about the impact this change could have in other scenarios. Exiting the virtual machine to the hypervisor is an expensive operation, and could introduce latencies if it was performed at the wrong time.

This is where the ecosystem shows its power. AMD technologies like AMD-V™ and RVI help reduce hypervisor and virtual machine overheads, allowing organizations to run bigger and more mission critical workloads. I think it’s a great demonstration of how hardware and software vendors can work together to solve a problem.

Much as the Musketeers learned, there may be obstacles along the journey, but perseverance and collaboration is what brings success in the end. The entire ecosystem benefits. And the results last a lot longer than a month.

Chris Barclay is responsible for product management and technology partnerships at Virtual Iron Software. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Any claims made herein have not been independently verified by AMD. 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.