High Availability and the Increased Need for Uptime

Systems and network uptime have never been more important for organizations. With a global economy that relies heavily on nonstop e-commerce, the appetite for downtime is dropping. Customers as well as business partners and employees expect continuous access to information and applications.

Because of these factors, ensuring high availability is — or certainly should be — a high priority for corporate IT.

“Downtime is no longer acceptable in many cases; it’s not an option,” says Robert Bready, research director at IT research firm Aberdeen Group.

Availability is proving to be critical as greater proportions of a company’s relationships with its customers depend on IT systems and services, says Eric Hanselman, chief analyst at 451 Research. “The toll-free hotline is going the way of the dodo, as mobile apps increase convenience and create tighter bindings with customers.”

The Importance of Being Available to Customers

As the financial services industry and other businesses have seen in the last year, availability is not just a matter of keeping applications running, Hanselman says. “The path to customers now needs to be resilient to denial-of-service attacks, adding a much larger element to the availability calculus.”

While availability is critical to customer perception, many enterprises have a difficult time attaching value to infrastructure investments to improve it, according to Hanselman. Only 13 percent of enterprises listed availability as one of their top two criteria for service providers in 451 Research’s most recent TheInfoPro study.

“This is a contradiction that organizations will have to address to meet customer expectations,” Hanselman says. Availability has to be a key component of application design, “and this extends from data-source management all the way out to service-provider selection.”

“While Active-Active application components and distributed databases are leading technology efforts, basic steps like having and practicing recovery plans are things that many enterprises have yet to accomplish. Capabilities don’t have to be technically sophisticated to be effective.”

What Downtime Costs Organizations

The cost of downtime has been steadily rising. Aberdeen Group surveyed 208 IT professionals in May 2013 and found that the average cost per hour of downtime across organizations of all sizes was more than $163,000 per hour. Large companies experienced financial losses of more than $600,000, medium companies more than $215,000 and small businesses more than $8,000.

These figures account for losses due to a drop in employee productivity and revenue, Bready says, but they don’t reflect factors such as tarnished corporate brand, reputation and customer attrition.

It’s not just a matter of systems being available; organizations need to ensure that users have fast, high-quality access to critical business applications. As a result, network latency has become increasingly intolerable for companies.

“User expectations of application performance have grown dramatically, even as more applications are delivered over challenging mobile environments,” Hanselman says. “Part of this is due to improvements in network infrastructure, but a growing portion is due to the increasing volume of multimedia traffic, where latency’s effects are much more visible.”

If high-speed access to critical applications is not available, “you may have 5, 10, 100 or 1,000 employees sitting around for an hour or a day,” Bready says. “If your applications are slow you’re not even going to know whether your customers are with you anymore or have left for a competitor. The appetite for latency is nonexistent.”

Strategies to Minimize the Damaging Effects of Downtime

In the event of downtime, how can a company minimize the loss of business and revenue?

“Businesses can minimize the effects of any outage through reasonable planning,” Hanselman says. “Ensuring that error messages give users not only an understanding of the problem but also a way to resolve it is a simple and crucial step in application design. Having a fallback plan will ensure that companies know how to react when failures occur.”

Having alternate, reduced-function websites and alternate providers can keep customers informed and connected, Hanselman says. “Companies should expect failures and practice responding to them.”

Organizations need to have comprehensive plans for both business continuity and disaster recovery, Bready says. And yet, a surprisingly high number of companies do not.

“We find that even though the cost of downtime is so high, and that one of the top pressures organizations are facing is the risk of business interruption, a lot of companies still don’t have a business-continuity plan in place,” Bready says. “We found that only 55 percent of all survey respondents have a disaster-recovery document in place.”

Disaster Recovery and Business Continuity Are One and the Same

No longer can companies just look at traditional data backup as a way to ensure uptime, Bready says. “In the old days, you would [perform] backups daily weekly or monthly, and if something went down, you would rebuild the server, reload the OS, applications and data and then resume operations.”

That doesn’t work in the current business environment, because it assumes downtime. “People have to think of disaster recovery and business continuity as one; they need to do both,” Bready says. “You need both practices to be in the forefront.”

Key components of any recovery and continuity strategy should include the ability to replicate physical and virtual servers for failover, Bready says. “If you’re involved in e-commerce, you need to replicate [systems] to other areas. Virtualization allows you to do that much more easily than with traditional servers.”

Continuous uptime is especially important for large companies in industries such as financial services and e-commerce. But even for smaller companies, in any business, being down for 24 hours can be costly.

“Financial services and online retailers get a lot of the focus in discussions on availability, but the need for greater resilience is just as important in a broad range of other industries,” Hanselman says. “Healthcare and manufacturing have clear requirements, as do any number of enterprises [that] need to maintain availability to provide value. Expectations from customers have grown, and astute businesses are looking to deliver value anytime that they want it.”

[image: Melpomenem/iStock/ThinkStockPhotos]