In today’s world the client experience is generally more about graphics while the server is more about I/O throughput. Of course there are times for both client and server when raw CPU power can be useful – but that is not the way you design either if you are focused on how the customer is going to use them.

Plenty of other blogs from me on clients so today, the 6^th birthday of the AMD Opteron™ processor, let’s talk servers.

I sold my first server back in 1986. It was an IBM 3090-120 mainframe (the 120 was a down-clocked 150 – see no idea is new!). The customer was an insurance company that was transitioning to IBM S/390 architecture for two reasons:  the application and the throughput on transactions the system got. Even back in those days, the design of a server was a mixture of CPU, memory and I/O.

Today, we face a dramatically changed landscape of “good enough” computing.  While performance still matters and always will, the real challenges customers face today are driven by restrictions on energy consumption and space . Simply put, we often find ourselves having more compute power than we can actually, well, power.

So as businesses around the world are now being forced to do more with less, they might want to take some advice from their IT department, where the focus on efficiency is now firmly entrenched.

As always, today AMD is announcing new products that help customers drive great value from their IT investments.  As we celebrate six years of the AMD Opteron processor we are also announcing our most energy efficient processor ever to hit the market – this processor is your ideal cloud computing platform.

There are five main design criteria behind all our server platforms:

·         Server utilization – the ability to do more work in the same physical platform. Our advances in virtualization and AMD-V™ technologies are good example.

·         Server performance – the ability to do more work in less time – paid off by the platform architecture know as Direct Connect.

·         Performance Density – the ability to do more work in less space, through hardware integration and by keeping the socket the same for a long time

·         Server Efficiency – the ability to do more work using less power. We have combined our current and some new capabilities into a power solution called AMD-P.

·         Value for money – the ability to maintain and in some places lead in price performance at most if not all value points.

We are seeing the server market increasingly defined by two main sectors – the 4P and high-end 2P x86 market, and the 1 to 2P very low-power market.  Despite the obvious differences the common thread is the architecture. And today we talked about the next wave of innovation on that front with our Direct Connect Architecture 2.0 .

But what about raw power I hear you asking? What about the pure 0-60 speed of your processors? Don’t we care about that?

Yes. But servers are also about data and throughput, and AMD has been a leader in X86 server design because we understand this difference.

You see, we understand that old mainframe joke, the one that goes like this:

Question: What do all computers do at the same speed?

Answer: They wait for data.

(You can laugh now)

Nigel Dessau is senior vice president and chief marketing officer 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 links sites and no endorsement is implied.