By Jason Waxman, General Manager, High Density Computing, Intel Corporation
I am not a fan of buzzwords. They take a perfectly useful concept and make them sound ridiculous. And in the tech industry we have a great propensity for coming up with a credible technology and assigning a name that makes it sound like pure hype. You don’t hear folks in the auto industry unveil the next great concept car as a new category called the “zoom transporter”, but that seems like what we’ve done with cloud computing. The reality is that there is both technical and market credibility behind the trend of cloud computing (and perhaps there is no better name). According to a 2009 IDC report, global IT spending on cloud is expected to grow from $16B (2008) to $42B (2012). One of the key factors resulting in this growth is the availability of cloud based services – both consumer as well as software as a service targeted toward business.
From my standpoint, the core element that makes something a “cloud” is the architecture of the underlying compute, network and storage in the datacenter. In the BC era (Before Cloud), we used to call the architecture “resource pooling” – the ability to dynamically provision resources based on requirements. Some of the leaders in driving the datacenter model have been service providers. What we’ve found from working closely with these companies over the last few years is that their requirements for optimization take on a whole new level. And with the growth of this infrastructure expected to grow at a 25%+ growth rate over the next 3 years, we need to spend more time thinking about how we enable this next generation of services.
From end user perspective, cloud-based services need to provide the illusion of infinite resources, sense of ownership and security and dictate a pay-as-you model. This in turn requires the data center to have high scalability, lower acquisition and operating cost, rapid deployment and support for multi-tenancy. For cloud-based services, the datacenter is the “cost of goods sold”.
At Intel, we’ve spent the last year thinking about what we do to enable these service providers today and what we need to do to enable the next generation of services. I want to share our thinking about intelligent power management, high density platforms, scalable management and software and datacenter optimization. We’re planning to announce new products and technology directions to support cloud services at the Intel Developer Forum (IDF) happening this week, Sept 22-24 at Moscone Center in San Francisco.
My session “Intel and Cloud Computing: An Overview of Intel’s Vision on the Cloud (PDCS001)” (I acknowledge the title feeds off the hype) at IDF covered the key technologies we see as fundamental to developers looking to deliver products for service providers and large datacenter customers.
Please share your thoughts and ideas on what do you think about this…