Key Cloud Computing Cost Considerations

The cloud computing vision is compelling. People quickly grasp the concepts involved and the potential for considerable cost savings. Some jump in too quickly, however, committing to platforms that don't fit their needs because they don't fully grasp all the variables at play. A more tempered approach to examining value and cost trade-offs is in order.

Organizations big and small need to consider the total cost of ownership (TCO) factors involved, and examine their own assets and requirements before diving in head first.

Five Less Obvious Implications as You Assess TCO

Hardware and infrastructure savings typically head the list of immediate cost reductions your company can gain from a move to cloud computing, especially platform-as-a-service (PaaS). Off-premise cloud applications cost less to maintain, patch, upgrade, and support.

The cloud is great for your team's career development. It might seem counterintuitive at first, but the cloud presents a number of opportunities for your internal IT staff. The staff members previously engaged in maintenance activities can transition to more strategic opportunities such as the development of new products and markets for the business instead of working to keep the lights on.

Though difficult to quantify, these shifts in staff priorities often matter more than simple "bean counter" savings. The effects and dividends become clearer as time marches on, and as the organization becomes smarter and more agile. There's a potential morale benefit as well, as IT staffers stuck with the mundane tasks of infrastructure maintenance get to lighten that load and show off their skills.

Scale out when you need it, even for one hour. The on-demand nature of cloud computing offers savings opportunities as well. Organizations simply pay for what they use. There are plenty of opportunities to save money when processing, memory, bandwidth, and storage can be dialed up and down at will. This applies to provisioning new resources, scaling back resources, and adjusting to seasonal and weekly demand spikes.

Domino's Pizza will be doing this when pizza sales go through the roof on Super Bowl Sunday. It doesn't buy a bunch of servers to cover the spike, and then sit on them during the rest of the year. Domino's will leverage cloud resources and dial up capacity when they need it.

Scale fast, and fail fast. The cloud also allows businesses to test new applications, projects, and business ideas quickly and easily without incurring hardware costs.

For many businesses, information architectures and unique process advantages that live in the software realm rather than the physical world have become their true "product differentiators."

A means to balance CapEx and OpEx. Finally, cloud computing allows organizations to move capital expenses over to the operational expense column on the balance sheet. Subscription models and the avoidance of on-premise hardware allow for this fundamental shift in accounting.

An agent to help with compliance. Many enterprises often have hundreds, in some cases thousands, of departmental line-of-business (LOB) applications serving business units and regional IT organizations. These applications often keep IT managers awake at night as they often lack centralized management, policy compliance, and hardware security — all of which can be necessary to keeping in compliance with any number of government and industry regulations.

A PaaS solution can provide an enterprise class on-demand computing environment with the flexibility the business groups need and the security and reliability centralized IT demands, without saddling IT (or the LOB owner) with the additional responsibility for maintenance and upgrades. Cloud offerings that fall under Infrastructure as a Service (IaaS) can also help with compliance because the owner has control over maintenance, patches, and upgrades.

That's the high-level view of potential cost savings. So what are the important considerations at play when planning moves to the cloud?

Drilling Down Further

If you're trying to figure out the TCO for moving an application to the cloud you need to understand existing business realities.

Compare your own operational efficiency to that of a cloud provider. The public cloud providers run highly optimized operations. Knowing if this is right for you requires a basis of comparison. That can take some due diligence into areas you may not be auditing regularly, but will be worthwhile given that cloud computing is here for the long term.

Consider such factors as:

• How much is being spent on data center power and cooling costs?
• What are the utilization rates for on-premise servers?
• How much time is the staff spending on maintenance tasks?

These questions need to be answered in detail.

Next, application usage and viability need to be examined. In terms of usage, each of the following needs to be quantified:

• The numbers of users of the application
• Does the application see demand spikes at certain times?
• Is there seasonality to the demand for the application?
• What is the expected rate of user growth for the application?

Knowledge of specific application growth expectations and demand fluctuations will prove critical when and if the organization transitions applications to the cloud.

Translate on-premise to cloud needs. The number of servers and virtual machines required for specific applications needs to be pinpointed. Can legacy applications be extended to the cloud? Are there viable alternatives to replacing legacy apps with newer, better solutions? Do some applications need to stay local because they're dinosaurs or because connectivity bottlenecks will slow performance? Application integration needs to be examined as well.

What about security and compliance concerns? Because of industry or government regulations the type of data you move to the cloud and how it's secured once it's there, needs to be closely examined.

IaaS or PaaS: Do you want to manage apps AND infrastructure; or just manage apps. Once the key data points about usage and transition viability are gathered, cloud provider capabilities have to be examined. IaaS gives you access to the infrastructure you want and keeps you in control. PaaS gives you the additional infrastructure you want but offloads the maintenance, patches, and upgrades to the provider (your staff focuses only on the solutions it is responsible to deliver).

You'll also want to examine if the platform is restricted in some way. Does it limit the type of software or operating system your application uses? What about development languages or tools?

The Microsoft® Windows® Azure™ Platform for example, is designed for broad interoperability, with support for multiples languages (.NET, C#, Visual Basic, C++, PHP, Ruby, Python, or Java), development tools (Visual Studio® or Eclipse) and Internet protocols (HTTP, XML, SOAP, and REST). Google App Engine, on the other hand, requires some Python knowledge, so Python developers would have to be on staff, added to the staff, or trained. The associated costs for maintaining that staff should be assessed as part of the entire evaluation.

Assess Your Cloud TCO

The Windows Azure Platform offers a TCO calculator for examining all these issues and planning for cloud transition. It will ask you many of the questions we discussed earlier in this article. By entering the details about the applications you're thinking about moving to the cloud, such as the amount of usage, the number of connections, the type of application, and more, the calculator returns a detailed report that examines everything you need to know about moving that applications to the Windows Azure Platform.

The Windows Azure TCO calculator returns an estimate of Windows Azure Platform costs over a three-year period based on the information that you provide (see Figure 1).

Figure 1. The Microsoft Windows Azure TCO calculator translates on-prem needs into cloud requirements, and delivers a customized report showing what your organization can save when it moves an application to the cloud.

The tool has three useful components for anyone looking seriously at the cloud:

1. The tool maps existing on-premise assets into what you will need in the cloud. Organizations can input a wide range of existing requirements, for example, and see exactly which applications and infrastructures can be migrated.
2. Collaboration features allow the sharing and adjustment of inputs, which is useful if you're working with a client or need to engage across teams.
3. One simple, customized document summarizes the findings that you can share with others in your organization. Specific Azure costs are estimated in detail, and the tool displays Azure costs alongside existing, in-house costs.

The TCO calculator is worth a look because there's a lot at stake with cloud computing in terms of both cost reductions and strategic agility. The advantages are not relegated to large organizations either. Cloud computing levels the playing field. Small organizations have the same opportunities to access IT services and talent that may have been out of their reach previously. New capabilities and innovation potential are within reach, and the costs can be easily quantified.

* This article was commissioned by and prepared for Microsoft Corporation. This document is for informational purposes only. MICROSOFT MAKES NO WARRANTIES, EXPRESS OR IMPLIED, IN THIS SUMMARY.