I recently read about a woman in Iowa trying to push for a new cloud type, and boy is the photo amazing. I was surprised when I read that we haven’t added a new cloud since 1951. But, maybe that’s because I work in IT, and lately it seems we’ve been adding new cloud types for data centers at a rapid pace.

While we talk about cloud computing as a new data center paradigm, in reality we have been steadily making the Internet a central communications vehicle for some time now. So, as the idea of cloud computing – the delivery of applications, data, and services via the Internet – gains more formal adoption, it is interesting to explore the types of clouds we are using today, and the technology driving these clouds.

One of the biggest topics of conversation I’m hearing is about internal vs. external clouds (I prefer this term over public vs. private because not all external clouds are open to the “public”). What we are really defining here is who “owns” the cloud, or, in other words, who owns the infrastructure.  Here’s how I see it:

• Internal Cloud – you are using the Internet as to deliver applications, data, and services to your users AND you own and maintain the cloud infrastructure (hardware, network, storage, OS, Hypervisor). In essence, you hold the power to say “hey, you, get off of MY cloud.”
• External Cloud – you are using the Internet to obtain services such as compute time, storage, and application stacks from a 3^rd party provider who owns the infrastructure.

I think it is actually more insightful to look at another side of cloud computing – one that classifies compute clouds by the type of services delivered as opposed to who owns the cloud computing infrastructure. So – here goes a look at the clouds in my life:

• Personal Clouds – providing services that help keep my life in order and the home fires burning.
  • Personal Communication Clouds – email and social network sites like Facebook for keeping up with friends and family
  • Banking cloud – where I go to pay bills and check on whatever money I have left after the current economic meltdown
  • Entertainment clouds – the TV guide and movies on-demand by my cable provider and the Wii network, for downloading my type of games (trivia point – AMD graphics processors power the Wii).
  • Shopping clouds – from big catalog companies to the small specialty business where I buy Kona Coffee
• Professional Clouds – clouds that drive AMD’s business or my professional persona
  • Business Communications Clouds – email, Twitter (you can find the professional me as @margaretjlewis), or Yammer (communications within the AMD community)
  • Company business clouds – Salesforce.com and AMD’s internal SAP site for example
  • Business communities – clouds where I hook up with other business people, like LinkedIn
• Performance clouds – an emerging type of cloud based on streaming data.
  • Gaming clouds – I must admit I am not an on-line gamer – so these are “foreign clouds” to me
  • GPS clouds – using spatial data to figure out where I am, like On-Star or any of the variety of GPS environments
  • HPC clouds – computational computer grids used by researchers and scientist to help solve today’s complex problems

So what is behind all of these clouds? A lot of sophisticated infrastructure – large server farms, high end networks, and petabytes of storage. A bit overwhelming. But once again let’s take a different look at cloud computing by keying in on one of the most basic component of any cloud – the processor.

The processor provides the computational and power to drive cloud clusters. It could be clusters of CPUs handling high rates of transactions like for an on-line catalog or clusters of CPUs and GPUs combined to provide astounding levels of computation for on-line gaming. One thing is for sure:  with the growing size web and cloud clusters, there is the need to balance performance with power consumption and space restrictions. This is driving demand for denser, more power efficient processors.

And, trust me: at AMD, we see how the world is evolving and take all of those factors into account. If you follow AMD, then surely you know all about our recently announced Six-Core AMD Opteron^TM processor (code named “Istanbul”). We are extremely excited about the potential of six-core performance delivered in the same power and thermal envelopes as our Quad-Core AMD Opteron processors, and so should everyone who cares about cloud computing. We are designing and driving platforms that offer a balance of performance, power, and price.

We are already seeing keen interest among cloud providers and web hosting companies in the Six-Core AMD Opteron processor’s capabilities to deliver more performance (more cores) in the same power envelope as Quad-Core AMD Opteron processors (code named “Shanghai”), and the potential for a total cost advantage. We see up to 34% more performance-per-watt delivered in the same socket infrastructure. And, we already have a number of major cloud providers using low power “Shanghai” (HE and EE) processors – so the stage is set for an upgrade to “Istanbul” – particularly when its low power versions are expected to be available later this year.

 The world of computing is changing – raw processor performance is being replaced with balanced system design. “Shanghai” continues to deliver performance/watt and price/performance and “Istanbul” is now joining the party.

 So, which clouds do you have in your life? There is a good chance that some of these clouds are powered by AMD processors.

Margaret Lewis, director, AMD commercial software and solutions

Margaret Lewis (@margaretjlewis) is director of commercial software and solutions 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.

@margaretjlewis