Business Continuity Goes Beyond Technology

Downtime can cost a company millions. In the United States, for example, downtime would cost a large brokerage firm US$6.5 million each hour.

Although technology is of paramount importance in keeping data availability high during major business interruptions caused by power failures, capacity glitches or natural disasters, attention should be also placed on the physical location of a data center and the infrastructure available, said Rob Kelly, managing director for Asia Pacific, Global Switch – an operator of carrier-neutral technical real-estate that manages 265,000 sq meters of secure and resilient space in eight facilities worldwide, including one each in Singapore and Sydney.

Data centers can be classified into three categories – prime (also known as main), mirror and disaster recovery sites. Depending on the nature of the organization, most of the prime sites are located in-house. Mirror and disaster recovery sites are hosted off-site, in a leased, outsourced scenario.

Infrastructure factors affecting a data center include air-conditioning in the building (for consistent cooling of equipment and to maintain a controlled environment); communications infrastructure (connectivity); physical and logical security measures (number of levels of security, if the building is not exclusively for data center use); building construction information (if the building can sustain the weight of generator, and other mission critical infrastructure);
fire protection systems (VESDA and gas suppressions systems); and reliance on water and power (if the building is not exclusively for data center usage, or if location is prone to outages).

“Business continuance solutions are in vogue today as seen in the numerous offerings by technology companies. Organizations should, however, go beyond these solutions and pricing per square meter and examine the physical location of their data centers and the infrastructure available at those sites as part of the overall business continuity plan,” Kelly commented.

In Asia, Singapore has one of the most reliable infrastructure and issues such as communications infrastructure and power are easily taken for granted, said Kelly. Elsewhere in the region, connectivity and natural disasters play a significant role in the decision process for data center location. “Nevertheless, organizations must evaluate their data center infrastructures or service level agreements with outsourced partners, to ensure that the sites can cope in the event of a failure.”

Kelly then cited an example based on the recent power outage experienced in Singapore. According to a report issued by Singapore power regulator, the Energy Market Authority, a computer malfunction caused the blackout that paralyzed parts of the country for 90 minutes on August 5, 2002. The blackout was the worst experienced by Singapore in ten years.

Global Switch’s Singapore facility, located in Tai Seng, was affected during this period and lost one of its incoming power supply sources. This loss of power was detected by uninterruptible power supplies (UPS) and prompted the building to switch to generator-mode for the entire 90-minute duration of the power outage. As a result, none of the customers in the facility suffered any measurable loss in power. When power was eventually restored, the generators continued running for 10 minutes to ensure that the power was not transient.

“The foundation of a data center, whether it is built or leased, is its infrastructure. Leased data centers may not be up for renewal in the next twelve months. However, now is a good time as any to review current data center location and infrastructure and determine if it is the best fit for the organization,” said Kelly.

News Around the Web