It used to be that large corporations were the biggest boosters of new and cool technologies; more recently the consumer has been driving technology trends. But with the timing of an economic recovery still uncertain and VC dollars hard to come by, it may be the Feds’ turn to set the stage — and technologies that fall under the “cloud computing” moniker are likely to benefit. The federal government is open to the idea of using commercial cloud computing solutions, according to Vivek Kundra, the country’s chief information officer.
In an interview earlier this month, Kundra told me that one of his initiatives has involved pushing for “the adoption of cloud computing and consumer technologies within the public sector.” While our conversation was wide-ranging, in the end we ended up focusing primarily on his belief that cloud computing can help the government be smarter and more efficient. For example, when I asked him what he planned to do about tgovernment-owned, energy-guzzling data centers, Kundra said, “What we have to ask is even a deeper question, which is does the U.S. government need to own all these data centers or is there a new computing model that we could leverage?” By way of context, as of January 2008, the U.S. government was using 1.2 billion kilowatts a year to run 600,000 servers.
Kundra thinks there are lot of technologies available in the consumer marketplace that it can leverage. “As I’ve looked at the processes and investments that we’re making across the government, a lot of the investments that we’re making make no sense,” he said. To illustrate his point, he shared an anecdote about the TSA wanting to set up a blog and how its approach would have cost the agency (and the taxpayers) about $600,000. Thankfully wiser minds prevailed. Saving tax dollars underpins his approach to all IT. When I asked him whether the Feds would opt for commercial services such as the ones offered by Amazon and soon Microsoft, or if they would they build their own clouds, he said, “It’s going to be both because on the one hand we can leverage platforms that are available right now in the market for information that’s not classified or sensitive in nature.”
The government has already laid out its plans for some of its pilot projects and from my conversation with Kundra, it was clear that he’s not averse to working with commercial cloud service providers. The General Services Administration also recently released a Request for Information (RFI) for Infrastructure-as-a-Service (IaaS) offerings.
Kundra and his team are working on launching a “digital storefront for cloud computing solutions” that would offer for federal and other government buyers an experience that “is the same or similar to the experience that you and I have in our personal lives when we go to online stores like amazon or eBay or any of the other online vendors.”
This move to the cloud is going to be good news for Amazon, RackSpace and Microsoft, all of which I think are going to be competing with with traditional suppliers of information technology to the government. Others say that Terremark, HP, Google, Cisco and various systems integrators could be among the big winners. (GigaOM Pro, our $79-a-year subscription-only research service, has started following these developments closely.) For the U.S. government, cloud computing could be an easy way to deal with urgent and important issues, such as upgrading the federal and state technology infrastructure without costly upgrades.
If Kundra is successful in his efforts, then the sheer buying power of the government is going help cloud computing reach a tipping point.
This may bode well for the government cloud but consumers will never reach that tipping point particularly with big brother leading the way. Personally (holding up hand,) No thanks.
Cloud computing is another technology that is more hype than reality. The truth is that most people don’t want it.
You can tell when people want a new technology because the snatch it up when it is still crude and keep buying any little improvement on it. For example, coloured graphics: back when it was just 8 colours and blocky, people thought it was wonderful. And when it went to 16 colours, people bought new computers just for the graphics. Even today when there is not much difference between them, consideration of the graphics card is a major part of the computer-buying decision.
Not so with cloud computing. If it is as wonderful as its proponents claim, people would have already move en masse over to it. The reason why anyone is still talking about it is that some software companies see it as a way to lock-in the
suckerscustomers. Getting the government to back a technology that few of the public wants is just a waste of money and effort.Shawn – you’re old man. People search ‘en masse’ with Google – cloud computing. People make purchases via Amazon ‘en masse’ – cloud computing. Salesforce.com is a billion dollar company – counting cloud customers and revenues ‘en masse’. Yahoo Mail and GMail are two of the most popular cloud computing email services in the world – no doubt, ‘en masse’. Gotten directions lately – Google Maps … hmm, maybe – cloud computing ‘en masse’. Vendor lock-in is no different from on premise – either case you would need to go through the steps of porting and normalizing data, business process, workflows, etc. Moving from Salesforce to another platform is no more difficult than moving from an Oracle environment to MSFT – since you don’t have to worry about infrastructure, the case can be made that it is even easier.
As a government, business, organization, or individual; cloud computing allows one to leverage the best services possible. One does not have to make large capital investments in software, hardware, maintenance, or upgrades – you pay for what you use, no more no less, and you are always on the most up to date version. With cloud computing any mom and pop shop has the same access to state of the art technologies as the largest enterprise. Want to end the recession – have every small business move from Exchange and Office to Google Apps.
Mr. Malik – I enjoyed your article, thank you. You seem to suggest that MSFT is not a traditional supplier. For the sake of competition hopefully Azure is a robust platform, but Amazon, Salesforce, Google and others are nearly a decade into their cloud computing adventure. MSFT’s 2008 Annual Report didn’t mention one specific Software+Services (note, not SaaS) product while the client/server cash cows, X-boxes, and Zunes are obviously areas of focus. MSFT could transform itself into cloud computing powerhouse, but for all their marketing talk they are currently still a very traditional, legacy provider with a very immature cloud offering that has yet to be tested in the marketplace. Just as Google’s announcement of Chrome OS does not make it a desktop player yet, the imminent arrival of Azure does not make MSFT a cloud computing company yet – ‘yet’ the key word in both instances.
Shawn, do a little research, you will find that organizations are moving to the cloud en masse. Merril Lynch predicts value of cloud marketplace to be > $160 Billion by 2011.
Gartner projects that Software as a Service (Saas) cloud computing will continue to grow at an annual compound growth rate of more than 22% in the same period.
In my personal view
Pro:
a) There is a business case if one has temp computing need over permanent/eternal.
b) Auto-scalability [if there is such thing] another sale-able point, where it is unpredictable on upper side that how much computing resources would be needed.
Con:
a) Not cost effective if one has permanent/eternal computing infrastructure needs. You may still outsource to traditional hosting companies, if you do not have internal resources. It is about choosing between traditional hosting (Data Center centric) and Cloud Computing on cost point of view. Traditional Hosting companies can provide scalable solution for a predictable needs.
b) One has to adapt thought process of Cloud Computing mindset, which is not easy or a has steep learning curve. It is not yet consumer friendly. That is not the case when choosing any traditional hosting company.
c) If you are thinking to migrate [any] existing application running in production environment – Cloud Computing is not a right choice. There is cost factor in transforming the system or application to specific Cloud Computing eco-system.
d) I do not think one can still move application/system between among various Clouds as all has their own distinct unique eco-system. And that is not the case when you use traditional hosting company. You can easily change camps without being captive.
If they can offload non-sensitive material like regulations, library of congress material, plain old informational websites, etc. (which is I suspect most of their material) , they can probably save millions — which could be invested elsewhere to upgrade the overall computing infrastructure. Having worked in the government many years ago i can tell you tehir hardware is generally years out of date.
I agree there’s no way you would ever want any sensitive info in the cloud.