The Journey to the Cloud

Cloud computing is a revolution at the tail end of an evolution.

It is similar to what occurred during the industrial revolution. Back in the late 1700s, an artisan might sit at a workbench and produce 100 chairs a year. But then, demand for chairs probably grew to about 1,000. The growing demand for chairs and how an artisan would need to scale up to meet the new demand changed how he built chairs. The innovation of the water wheel being connected to the mill became a competitive advantage. Eventually, when electricity became available and it revolutionized production, nobody cared about the watermill anymore. Computing is no different. It’s about the scale of delivery and demand.

About this Author

Patrick Zelten

Vice President, Forsythe IT-as-a-Service Practice

read profile | recent posts

What is Cloud Computing?

Cloud computing is a model where IT compute resources—be it hardware and/or software—are shared and accessed through the Internet and utilized on a subscription, pay-as-you-go basis.

The Evolution to Cloud Computing

Looking back, there has been a natural evolution to cloud computing. We witnessed the dawn of punch tabulators in the late 1800s, followed by calculators, desktop computers (mainframe and client servers), mobile computing and, finally, cloud. Cloud really evolved through Google, Amazon and Facebook because of the scale of supply and demand. Facebook, for example, has one billion users. When you go from 100 servers to one million servers, you simply have to run things differently. Demand and supply at scale changes everything. So cloud data centers evolved to keep up with demand.

Today, people have more computing power in their pockets than companies had in their data centers in the 1980s and 1990s. The current iPad is as fast as the supercomputer was in 1993. In the past, it made sense for companies to build their own data centers because computing was scarce. Early adopters purchased computers for their businesses, and they had a huge competitive advantage and experienced tremendous productivity gains. It made sense back then. But now computing is cheaper and accessible from almost anywhere.

The Cloud Changes Everything

There are two important concepts to think about when looking at cloud computing. The first is around economics. Cloud computing is a radical economical innovation. Prior to cloud computing, if a business needed, let’s say 2,000 hours of computing, it would buy a single server. Consider retail companies during the holiday season. They may buy 10,000 extra hours of computing during peak time, but it was never used again throughout the year because sales volume would drop back down to normal levels. Today, we live in a world where it costs the same to rent one computer for 1,000 hours as it does to rent 1,000 computers for one hour. We have the ability to correlate the demand for computing for cost computing. Computing used to be scarce, but it is not now. We are able to do more things affordably. This is a radical transformation.

Second, cloud computing has changed the way information technology (IT) is being consumed and delivered. Cloud computing plus cloud data centers equals cloud services. Those two things have offered customers a new option. Businesses used to put all of their data inside the four walls of a traditional data center. But now, businesses and the different departments within a business can take advantage of cloud data centers where a customer pays for the service based on how much they use it. It’s a new model that delivers a huge scale at a low cost.

Keep in mind, however, that IT is not any less accountable, especially when the application goes down or is unavailable. IT still provides and oversees the fundamentals, such as security and availability, and also optimizes IT services that physically reside inside and outside the organization.

The Future: The Hybrid Data Center

Now that we understand the big picture that surrounds the cloud, the reality going forward for most medium and large enterprises is truly a hybrid infrastructure, which includes management of internal and external infrastructure and applications. Such enterprises have the usual constraints with legacy applications and embedded infrastructure and data centers, not to mention data security and recoverability concerns. On the other hand, small companies and start-ups typically do not have such constraints and can adopt a more exclusive cloud model more readily.

Therefore, it is best to look at the cloud as just one of several delivery models that exist to support different cost, control, service level, and risk constraints. For example, internally, a company may be managing a legacy environment, converged technologies and/or private clouds. And externally, a company might have a colocation environment, private clouds, public clouds and/ or outsourced components.

Companies can make decisions now about housing different types of data internally or externally. What offers the most competitive advantage? For example, some might decide to use a cloud provider for email and customer relationship management tools from salesforce.com, but human resources and accounting data will stay with legacy systems in-house.

Cloud computing is typically the most appropriate delivery model for applications or workloads that are virtualized, require rapid deployments and have seasonal or quick processing demands, such as retail businesses during the holiday season.

Some of the overall key benefits of the cloud delivery model include:

Speed to Market. How quickly do I want to be able to react to business demand? Whether your enterprise is small, medium or large, one of the key benefits that the cloud offers is speed to market. Companies can quickly react to business or demands by having the IT infrastructure in place to rapidly roll out new products or applications, such as a customer relationship management (CRM) system for a sales team that quickly wants to grow in a new market.

Businesses can open in new markets or geographies without worrying about the time it would take to set up a physical data center.

Elasticity, Agility and Scalability. How quickly do I want to be able to transform my infrastructure? Scalability is the key here. Businesses have access to infinite resources whenever they need it. Provisioning data in the cloud for a company can be done in hours, versus a physical data center implementation that could take months. Using the cloud means that a company can react, purchase, and deploy exactly what it needs exactly when it needs it. Consider Groupon, for example. This young company, which features a daily deal on “the best stuff to do, see, eat, and buy,” is already in 43 countries. Much of their success is due to their fast go-to-market strategy and their ability to use the cloud computing model.

Only Pay for What You Use. Cloud computing is a pay-as-you-go approach. Most cloud providers will require a low initial investment to get started and then an additional investment as system use increases. In companies where expense is favored over a capital outlay, this pay-for-what-you-use model is valued by not only the chief information officer (CIO), but the chief financial officer (CFO) as well.

Simplifies Operational Management for IT. The cloud reduces the burden of ongoing maintenance and staffing costs because it can automate mundane IT tasks, which frees up IT staff to focus on innovative technology, strategy or product development.

A Possible Reduction of Total Cost of Ownership (TCO). IT staff can spend too much time maintaining the current environment. Or the business might be struggling to meet the rising costs of maintaining that environment. It is worth checking into outsourcing to a cloud provider, as they may be able to provide a reduced TCO.

Planning Your Success

So now that you understand the value of cloud computing and are ready to possibly use it as a delivery model, how do you practically get started? Most companies that venture into the cloud begin with a proof of concept or a pilot first. As you prepare your pilot, try to keep these five points in mind.

1. Make sure your fundamentals are in good working order.

Virtualization, standardization, consolidation, security, IT service management processes, and business continuity and disaster recovery are still very important. Make sure these fundamentals are considered in the planning process. Also, don’t abdicate control or responsibility for those things just because an application or applications have been moved to the cloud.

2. Differentiate applications.

Each application is different. Security is an important characteristic, but it doesn’t apply equally to everything, as not each application needs the same level of security. Know the application and its requirements, and then apply criteria so that you can make a decision. Also, define your strategy around two major points: value and investment. Businesses can quantify and rank or rate high-level benefits and the risks of moving and effort on an XY axis. Regarding value, we are talking about tangible and intangible benefits and benefits from speed and flexibility. For investment, we refer to the effort needed to get to “cloud ready,” which includes cost and risk mitigation efforts around virtualization, disaster recovery, and security. Functions that fall into the cautionary category (e.g., low value and low investment) are actually great places to start with a pilot because expectations of these benefits are low and the focus can be on operationalizing the cloud, not on achieving immediate benefits.

3. Start small.

An important point about cloud—it is not necessary to devise a fully baked strategy before dipping your toe in the water. The upfront investment and the cost to change are not significant enough to warrant an exhaustive strategy. When an organization moves from proof of concept to broader adoption, a strategy is necessary. It helps to ensure the cloud is being used appropriately to meet the needs of the business for cost effectiveness and to mitigate risk. Be sure to take the knowledge from performing a proof of concept and build them into your overall cloud strategy.

4. Select a service provider that meets your needs.

On the surface, it looks like all cloud providers have similar capabilities. However, they certainly differ in the full breadth of their offerings, their underlying architectures and their contractual approaches. Are you looking for both colocation and cloud from a single provider? Are they building their cloud on technology that is scalable, supportable and consistent with your own environment? Are they providing true pay-as-you-go costing or a term contract with minimum levels of usage? These are just some of the questions to consider during a provider evaluation.

5. Identify your barriers.

Most of the challenges with this new technology (remember the beginnings of mobile phones) can be overcome. The biggest concerns about moving to the cloud include availability, security and compliance, recovery, lack of control, management and monitoring integration, and lack of internal expertise. Security and lack of control top the list, but both can be addressed and managed to a comfortable level that allows companies to move forward with at least a portion of their applications or workload.

Your Journey to the Cloud

As the paradigm shifts toward this new technology, remember that cloud computing is not an all-in bet. It is not a right fit for every circumstance. Ultimately, a CEO or CFO will ask how the internal IT organization measures up against provisioning, utilization and power, and cost benchmarks readily available by cloud service providers. As IT leaders, it will be important to know you stack up against these benchmarks. Being able to promote the cloud, where it makes sense for your business, and how to go about it will put your business on solid footing for continued success. Ultimately, if you have already started your journey to the cloud, it is the right place to be. By testing the waters with this new technology and delivery model, you can find out how your business can best reap its value.