To cloud or not to cloud? Is that the right question?
It’s not a question, but George Santayana once observed, “Those who cannot remember the past are condemned to repeat it.” Cloud represents tremendous opportunity, but it’s the latest of many paradigm shifts. Here’s a better question, “Can we learn from the past?”
Take a step back and consider what all of the previous game-changing technologies had in common. Whether you recall the client-server or n-tier architecture transition or the adoption of server virtualization – which is still largely ongoing, according to most analysts – the adoption of these new technologies and operating paradigms had many fundamental similarities.
Apologies to William Shakespeare aside, a recent web search for the phrase “to cloud or not to cloud” turns up over a dozen current articles on the latest challenges and opportunities in the cloud. To their credit, the authors all advise caution and offer many excellent recommendations on how to evaluate cloud providers, build a cloud business plan, and minimize risk to the business as you move forward with your adoption of cloud. Cloud promises new ways to control costs, offer innovative new services, and respond more quickly to new business opportunities. Those promises may even sound eerily familiar.
Each of these new technologies had to pass the economic sniff test before they met with wide adoption. Vendors offering game-changing technology offered bargain-basement prices to make the new technology broadly available as quickly as possible, but the broader market usually waited for early adopters to “skin their knees” on the technology first. Once that first wave of adopters found the worst of the bugs, enterprises looked to the vendors to offer best practices for adoption and for analysts to develop ROI models that described the lowest-risk adoption plan.
If you found this blog post – either through a web search, RSS feed, or linked site – you’re probably at least considering the ROI of cloud already, so let’s just assume for the sake of argument that the economics of cloud are somewhat compelling. It doesn’t matter whether the particular solution that entices you is SaaS-based CRM from Salesforce.com or IaaS from Amazon or something else. If the economics make sense, you have moved past window-shopping and now you have to consider how to reconcile yet another game-changing technology with the way you currently run your business.
Remember as you carry out your economic evaluation of cloud to consider the costs of tools and training. Tools require headcount to operate and training is an ongoing rather than a one-time expense, so if you buy tools that can manage the existing services as well as new cloud-based resources you’ll save money on the tools. If the processes that you employ to manage the current IT environment also encompass services you manage in the cloud, the implementation and management of those services will add little to the current cost of management.
No one ever made a wholesale “light switch” transition to new technology. At best, the transition was managed along asset refresh timelines. In other words, as the old platforms were depreciated or old OS and application versions were no longer supported, new gear was installed and the new paradigm was implemented. This enabled IT to sunset old applications with the hardware on which they ran and to implement new technology as it made business sense.
Can your business make a wholesale transition to cloud? Can you use SaaS-based CRM or expense reporting? Would private or public cloud-based VM hosting work for your developers? While it’s possible that one or more of these technologies would work for your business, it’s highly unlikely that you will be able to move everything. The more reasonable strategy – as recommended by all of the “to cloud or not to cloud” authors – is to choose the services that make sense to run in the cloud and migrate them. The implication of this perfectly reasonable strategy is that you will also need to adopt new tools and processes to manage these cloud-based services.
Tools and processes that encompass cloud-based services as well as the services that remain in your traditional data center will reduce the cost of migration as well as the operational costs of services in both environments.
Every new technology solution required new skills. New skills take time to acquire, whether through training or new talent acquisition. Even if you hire an expert off the street, it still takes time for that person to familiarize themselves with your business’ standard operating procedures and expectations. It also takes time to teach them how to work with all of the other legacy systems upon which they will continue to depend.
Remember that the skill gap is really two gaps. There’s the gap between existing employees’ skills and the skills they will need to work with new technology. There’s also the gap between newly-hired experts’ experience and your business’ policies and procedures, or “how things are done around here.” If your processes are well-ordered and well-documented, existing staff can be repurposed while some of the team members are trained on the new technology. Good processes also make it easy for new team members to come up to speed on how their efforts will complement the rest of the organization.
A transition to new technology – like cloud – offers the opportunity to establish a set of operational processes that govern how IT works with the existing infrastructure as well as the new technology. Look at proven process frameworks like ITIL as a way to standardize operations. Change management and software license management apply to the cloud as well as to the traditional IT environment.
What was the question again?
BMC’s vision for Business Service Management includes a single view of the physical, virtual, and cloud-based resources that IT uses to provide services to the business. When we decided to adapt to the cloud, we built BMC Cloud Lifecycle Management on the foundation of trusted cross-platform management solutions, including Remedy and BladeLogic. Change management processes that have been tried, tested, and validated in thousands of enterprises now extend seamlessly to the cloud. Provisioning processes that deploy physical and virtual machines in the datacenter now treat Amazon EC2 and other cloud providers as equally valid provisioning targets.
Thousands of IT organizations have learned the hard way that tools and processes that operate in siloes are neither flexible nor scalable.Use this opportunity to adopt processes that will enable your organization to adapt to technology changes, not just respond to them.
Cloud promises to change everything. Again. The question “To cloud or not to cloud?” may sound like a choice, but the real choice is whether or not we learn from the past.