As pressure mounts to deliver value with ever-increasing speed, lines of business (LOB) are often drawn to cloud computing's ease of use, flexibility and rapid time-to-value. The resultant Shadow IT created by use of consumer grade cloud computing resources usually raises questions about enterprise security, but the real risk is the potential for downtime due to inadequate availability.
Any interruption that impacts the customer experience will affect the bottom line — and a company's reputation — faster than you can say "temporarily unavailable."
So what can IT leaders do about it? With the cloud movement a foregone conclusion, how can they ensure the requisite availability standards are met — and that their investments in availability are the right ones? There are five key success factors for addressing Shadow IT while ensuring availability in the cloud:
1. Think about which applications belong where. The first step is to know which applications make sense to deploy in the cloud and what their availability requirements are. Let's face it, not every application requires "nine nines" of availability. A basic collaboration tool can probably go down for a few minutes without bringing the business to a crashing halt. These applications may be just fine living in the public cloud because they have no special requirements for security, availability, or compliance. Mission-critical applications are a- different story. Because these have strict requirements they are better suited to deployment in a private cloud, where you have greater control.
The goal is to balance the need for agility with the need to maintain appropriate levels of availability and security. Taking a "one size fits all" approach to availability can be extremely costly, while potentially leaving critical vulnerabilities. Analyzing your cost of downtime for specific apps, including cloud apps, is crucially important to ensure you're deploying them appropriately and making the right investments in availability.
2. Make availability understandable to the business. Ever feel like IT and the business are speaking different languages? Instead of getting frustrated that LOBs "don't get it," focus on helping business leaders understand how cloud availability impacts them. And listen carefully to what their needs are as they relate to flexibility and availability. Then work with LOBs to provide solutions that meet their business needs, while providing the level of availability required for specific applications. Here's where private clouds can be attractive, delivering the flexibility the business wants together with the control over availability that IT needs.
3. Read the fine print. With the rapid adoption of public cloud technologies there is no shortage of ambitious availability claims from mainstream vendors with familiar names. Unfortunately, many claims are more vapor than fact — at least in terms of delivering true, mission-critical availability. It's wise to read the fine print in any public cloud vendors' service-level agreements. You may be surprised to find that some do not classify continuous outages as "downtime" unless they last 5 minutes or more. That means your application could go up and down all day long but, as long as it was not down for longer than 5 minutes at a time, they count that as 100% uptime. And if they do have an outage, all they offer is a 5% credit on your next bill for the server that was out.
Sign up for CIO Asia eNewsletters.