Bob Evans| Forbes
While cloud computing has made huge strides in the past year toward becoming a mainstream business tool, lots of half-truths, mistruths, and good old-fashioned falsehoods are still swirling around with regard to the cloud’s place in the enterprise.
Quick quiz: did you know that HAL, the control-freak computer in “2001: A Space Odyssey,” was a cloud-based system?
Quick quiz #2: did you know that half of the digital watches in Asia are now cloud-compatible?
If you answered yes to either of those questions, then you really need to read the rest of this column because I just made up those two whoppers. And even if you answered no to just one—or even to both—you should still check out these 10 myths because of the speed at which cloud adoption is growing, and the business potential that the cloud now offers.
Seeing past these myths is critical to making the right decisions about whether, when, and how to adopt cloud-based solutions. So based on a keynote presentation I gave yesterday at CloudExpo New York, here’s our list of the Top 10 Cloud Computing Myths that we’d like to dispel:
Myth #1: Public cloud is the only true cloud. While public clouds continue to garner more notoriety than private clouds—particularly among the media and capital markets—the truth of the matter is that the volume of private cloud adoptions still exceeds that of public cloud adoption. In a 2012 Gartner Data Center Conference poll, almost 9 out of 10 survey respondents said they were planning, piloting, or already using a private cloud within their organizations. In addition, a growing number of businesses are deploying hybrid clouds that integrate data and business processes across both public and private clouds.
Myth #2: You’re either in the cloud or not. Since businesses today need to be able to respond rapidly and comprehensively to new trends and marketplace realities, it only makes sense that businesses are pursuing a wide range of cloud-based solutions to meet their specific needs. In addition, for many businesses, moving into the cloud and converting more systems and processes to the cloud requires an evolving and multi-step approach that adapts to constantly changing business needs and priorities. In many cases, the first step is to move from siloed environments to consolidated or virtualized environments. In other cases, companies move to private clouds by introducing self-service and auto-scaled environments, or metering and chargeback. To some, specialized public clouds provide a rapid way to adopt the cloud, particularly LOBs looking to quickly deploy SaaS-based solutions that addresses particular business functions. At Oracle, we believe most businesses will eventually move to hybrid clouds in which applications span both private and public clouds and are managed in a federated manner.
Myth #3: Clouds are one-size-fits-all. As some companies have learned to their great disappointment, there’s no such thing as a cookie-cutter solution when it comes to cloud—regardless of how frequently and loudly some cloud providers shout try to say otherwise. Today, organizations have a huge range of cloud options:
*Different deployment models: public, private, and hybrid;
*Service models: Software as a Service, Platform as a Service, and Infrastructure as a Service; and,
*Operating models for a cloud solution: the customer owns and operates it; the cloud provider owns and operates it; or, the customer owns the solution but the provider operates it.
The notion that businesses must be confined to a limited and closed set of cloud choices is as silly as saying they can only make one type of product or sell through a single distribution channel. As with every other type of business operation or asset, companies need and want to have a wide range of cloud-computing choices to match up with their various requirements.
But what about virtualization–isn’t that the same as the cloud? On to Myth #4:
Myth #4: There’s no difference between virtualization and the cloud. It’s true that virtualization is a key technology that enables cloud computing, but virtualization alone does not make a cloud. While virtualization is primarily focused on server and workload consolidation with the goal of reducing infrastructure costs, cloud computing involves much more than that. Consider that a recent survey of its members conducted by IOUG (Independent Oracle User Group) revealed that cloud customers are adopting Platform as a Service more rapidly than Infrastructure as a Service. That’s because those businesses are finding that while the infrastructure-centric approach taken by virtualization offers significant value, those companies are gaining even more agility and flexibility and savings from the benefits that Platform as a Service provides: higher levels of standardization and resource sharing, less heterogeneity and complexity, and lower costs.
Myth #5: Clouds only run on commodity components. While some well-known online consumer sites have gained some publicity for using cheap, commodity hardware, that approach is only one of many: clouds are running on servers of all different shapes and sizes. The critical issue in choosing the right type of infrastructure is not what some high-profile Internet company does, but rather what is the best approach for your business right here, right now? Whether you’re building a private cloud or consuming a public cloud, here are some important questions to consider:
*How much time and cost is involved in deploying hardware and VMs?
*Are you getting the lowest cost-per-performance possible?
*How does your quality of service (reliability, latency, availability) compare against other clouds?
As you dig into those questions, you’ll quickly find that commodity hardware doesn’t always provide the best value in terms of not just performance but also real estate required, power consumption, administration, and more. It may go against the current fad, but cloud hardware that is specifically optimized and engineered to deliver higher performance and efficiency often delivers the lowest TCO for cloud.
Myth #6: Cloud will lock you in. Without question, some public-cloud service providers would like nothing better than to lock you in to their proprietary cloud technology. Conversely, the good news is that businesses have lots of options that will allow them to avoid lock-in—provided those businesses know what to look for, including the following:
*Open, industry-based standards;
*Multiple application deployment options across public clouds, private clouds or on-premise; and,
*Seamless integration capabilities for moving data in and out of the cloud with ease.
Myth #7: Cloud is about pay-per-use. Paying for the use of a public cloud on a per minute or hourly basis may sound economical, but the costs are variable and can add up very quickly. Pay-per-use makes a lot of sense for applications that have a short lifespan or have large fluctuations in capacity needs. Subscription-based pricing that offers fixed monthly or annual costs are often the better choice for applications that are long-lived and have relatively stable capacity requirements. So before making your choice on public or private, consider what’s most economical for your particular application and weigh your options.
Myth #8: Public clouds are still not secure. Security has long been one of top concerns among organizations considering a move to public clouds. The fact is that businesses often improve application and data security by leveraging enterprise-grade public clouds. Many corporate data centers have limited security resources and expertise, challenges meeting regulatory requirements, outdated software and hardware, and don’t perform regular security audits and assessments. On the other hand, tight security is table stakes for any public cloud provider as many have the following:
*a dedicated team of cloud security experts,
*processes that ensure full compliance with regulatory, and industry standards,
*regularly scheduled third-party security audits, and,
*automatic updates for their hardware and software.
Still, not all cloud security is equal and the best advice is to review your cloud provider’s security technology and practices to understand any potential security risks.
That last sentence refers to a singular cloud provider–but won’t I need multiple cloud providers? Let’s take a look at Myth #9:
Myth #9: I need multiple clouds to run my business. There are enough SaaS, PaaS, and IaaS providers vying for your business to make your head spin—which is probably a good description of what your head’s doing these days. While it might seem like a good idea to select a wide range specialized cloud products and services to address different critical business functions, consider the data silos and fragmented business processes that you’ll face when deploying multiple specialized clouds and applications. You’ll have to deal with complex and costly integrations as you weave all those pieces together and try to maintain them, and you’ll be guaranteeing that your business users will face multiple, inconsistent user interfaces and experiences. A recent survey of business managers reported staff downtime, missed business deadlines, and stunted innovation initiatives due to poor integration of cloud applications from multiple vendors (source: independent market research conducted by Dynamic Markets in May 2013: Cloud for Business Managers: the Good, the Bad and the Ugly). A complete, integrated enterprise solution like Oracle Cloud eliminates the need for complex integrations and disjointed user experiences that result from using multiple specialized clouds. For more on this vital issue, please see our recent column called Cloud Computing And The Integration Quagmire.
Myth #10: The biggest benefit of cloud computing is lower costs. Reducing operational and capital costs are certainly good reasons for adopting cloud computing. For many organizations, the initial step in the evolution to cloud involves consolidation that drives significant cost savings. But these days, adopting cloud-based solutions is rarely just about saving money. Businesses are realizing that further evolution to a pre-defined service catalog, self-service provisioning, and auto-scaling drives speed of deployment, business innovation, and agility/adaptability to changing business needs. Others view cloud as a way to gain flexibility and to reduce the risk of failed deployments. The biggest benefit of cloud may differ from one organization to the next and is usually directly tied to whatever issue is driving cloud adoption in the first place.