Practice overtaking theory in cloud computing

It's getting harder to focus on the vision of cloud computing these days. While there are still plenty of critical and complex problems to solve, and many, many implications of this disruptive operations model that have yet to be understood, the truth is that we've entered a new phase in the evolution of cloud adoption. Real work now exceeds theory when it comes to both new online content and work produced.

This kind of snuck up on me, but it shouldn't have. I myself witnessed many of the early events that greased the skids for real cloud success: the introduction of revolutionary products from Salesforce.com and Amazon Web Services; great blogs that discussed practical applications of early cloud environments, followed by books that explained step-by-step what should be considered in application architectures destined for the cloud.

The rapid adoption of "software as a service"-style offerings from the likes of Salesforce.com, Google, Zoho, and a wide variety of others in both the consumer and business markets belied new computing options delivered at Internet scale.

However, what really made me aware of the changing cloud buzz is what's happening in the software development space. I was shaken awake by Microsoft's brilliant launch of its Azure cloud service. I loved almost everything about how Ray Ozzie and crew positioned and discussed Azure's services to its target market: developers of the next generation of business applications.

The recent (re)unveiling at Microsoft's Professional Developers Conference in Los Angeles included an impressive array of services, customer testimonials, and partner announcements. If it had stopped at that, I would have assumed it was just "Mister Softy's" massive marketing machine in action.

However, I began following the "#azure" tag on Twitter from that day forward, and I've been blown away by the amount of content being generated by developers for developers. For example, this step-by-step guide to installing SQL Server on Azure. Or, how about this list of sessions from PDC from a variety of vendor and customer presenters, covering topics ranging from development basics to "making sense out of ambient data".

But it's not just Microsoft. Other cloud platform and infrastructure service vendors are building significant volume. Ruby on Rails platform service vendor Heroku reportedly hosts more than 40,000 applications now. At their Dreamforce conference in San Francisco, Salesforce.com mentioned they had approximately 135,000 applications running on their Force.com platform. (Of course, the number of these respective applications that are generating revenue or even used on a regular basis was not disclosed. Still, these numbers are impressive.)

Amazon Web Services has seen tens of billions of objects stored in its S3 environment (64 billion as of August 2009), and reportedly has several hundred thousand instances running at any given time. Google App Engine doesn't seem to do much marketing, but anecdotal evidence suggests there is a large body of Web application developers running on both the Java and Python instances.

Development and test services, such as SkyTap and Soasta, are thriving. The cloud model really works well for the dynamic resource usage model of software engineering. In fact, it works so well that IBM is putting some real muscle into the game.

There is other evidence that cloud is seeping into mainstream IT thought. This year's Gartner Data Center conference has a "virtual track" dedicated to cloud computing and its impact on the data center. Several vendor conferences leaned heavily on cloud computing in the last year. Professional associations are getting into the act by considering the impact of the cloud on their respective best practices and standards.

There is growing evidence that new and existing independent software vendors and consultancies are finding the cloud to be fertile ground. Of course, that could be a double-edged sword, as some firms will try to use the cloud as leverage to pry their way into otherwise closed doors. However, real projects do exist, and there are signs that that opportunity is growing.

If you are wondering if cloud computing is a fad, the evidence to the contrary is all around you. I heartily recommend that you really listen to what is being said, understand how the cloud is being used, and seriously evaluate how this disruptive model will change your projects, your organization, and even your career. Clearly, there are many technologists who already have.

[t8]

Hi, I am coming from the Cloud. What, you are still running Windows and Windows servers.<br />I think I will float away and come back when you are more advanced.

[wattersjames]

What a great array of examples here of meaningful cloud deployments. Every-day I try to remind our partners and customers how much 'commercial' activity is already happening in cloud-computing. <br /><br />Azure is certainly a massive, massive capital and research investment from Microsoft that is hard to over-state in its long term strategic impact on how people consume MS IP. <br /><br />I'd also ad Engine Yard to your list. They haven't said much publicly--but I've seen the size of some of their big app deployments and they are 'massive.' They would not have gotten 34M$ in new venture funding without that very real revenue run rate.

[Gpadakandla]

Great coverage on Cloud service offering and rapid adoption. However, the myriad of vendor offerings, each with their own pros and cons, further complicates the Cloud vendor selection process for the customers. To get true "bang for their buck" business customers have imminent need to develop innate understating of various cloud offerings, associated total costs/risks and limitations. <br /> <br />- Gopi Padakandla <br />Blog: http://www.enterprise20link.com/

[prestondeguise]

And yet many of these examples also should come with large warnings as to service outages that users have experienced in the past and a lack thereof of any tangible promise of data protection by the service providers. Cloud Hype insists that Cloud is representative of Disruptive Technologies. Maybe, but not disruptive as a life altering technology, but disruptive as in "Why can't I access my data?" and "What do you mean my data is gone?"<br /><br />I find it appalling that Cloud proponents are rushing to accept Azure is some fantastic offering after the entire Sidekick debacle without in any way requiring Microsoft to make guarantees of better service delivery than that provided by Danger, a company wholly owned/controlled by them.<br /><br />There is nothing significantly new or disruptive (in a good way) about the Cloud, and it would be beneficial if everyone pulled back, took a couple of breaths out of a brown paper bag and actually started approaching it logically and cautiously.<br /><br />Preston de Guise<br />Blog: http://iamtheanticloud.wordpress.com

[jpanagulias]

I have to respectfully disagree with Monsieur de Guise and side wholeheartedly with James.<br /><br />The evidence points to the viability of cloud computing, not its demise. Cloud computing is part of an ongoing evolution of computing. To say, "there is nothing significantly new or disruptive" is as much overblown as those that say it will change life forever, starting...now. It will be a game changer over a period of time.<br /><br />Simon Wardley of Canonical did a nice job of eloquently explaining cloud computing as part of a social and technology evolution. (Check out his video &lt;a href="http://www.youtube.com/watch?v=okqLxzWS5R4&gt;here&lt;/a&gt;. <br /><br />If you consider Salesforce.com, then you can even argue that cloud computing has already both been around for some time (the company is 10 years old) and it has positively impacted businesses both large and small. (The company has over 67,000 customers and revenue over $1 billion).<br /><br />Similarly, look at the companies James mentions. Soasta is a great example of a cloud computing provider that fills a very viable corner of the IT market: performance testing. There are many more examples where cloud-based vendors tackle interesting areas that are ripe for re-invention, whether around on-demand, pay-as-you-go servers and storage (Amazon, GoGrid, Nivanix) or those looking at the very issues people bring up time-and-again as "deal-breakers" for cloud computing: security, vendor lock-in, etc. For example, CloudSwitch's solution allows a company to move their existing applications to the cloud without re-architecting the application or changing management tools and policies.<br /><br />Of course, cloud computing is not without its issues. But, to argue that it is all hype misses the reality that real, tangible cloud-based services are changing the way businesses look at IT.<br /><br />John P<br />&lt;a href="http://cloud.kendallsquare.com"&gt;cloud.kendallsquare.com&lt;/a&gt;

[TedNewman]

The growing reality is that many organizations expect IT to be delivered as a service that adds business value: Business agility depends on IT agility. Our customers want to cut through the hype of cloud computing and hear about the practical, pragmatic steps they need to take to deploy a private cloud that will provide measurable business results such as reduced time to market for their products. Our steps for success include developing a business case, building a scaled out virtualization architecture, and implementing a management toolset that provides a single view into IT operations. <br /> <br />Ted Newman- EMC Consulting <br />blog: http://mrinfrastructure.com/

[ocads]

Due to a customer request I did a quick comparison on how cost effective it is to move our small business contact manager product to the cloud environment. I found that with the current pricing it will be difficult fo convince small business owners to switch from the dedicated to cloud environment. Here is the comparison: <br /> <br />http://blog.officeclip.com/post/2009/12/02/Cloud-Computing-Is-Dedicated-Hosting-still-Cheaper.aspx