Disaster recovery programs are failing because of a lack of resources and an increase in the use of virtualisation, according to research.

More than a third of disaster recovery (DR) tests in Australia have failed to meet recovery times, according to the 2008 Symantec Disaster Recovery Research report which polled 1000 IT managers across 15 countries during June and July, while 16 percent are successful.

The figures are exacerbated by a lack of DR testing by local businesses, according to figures. Up to 47 percent of respondents stated they run DR tests no more than once a year because of poor resources and the disruption it causes to the business.

Symantec vice president of the Veritas cluster server group, Mark Lohmeyer, said virtualisation is a thorn for DR testing.

“Virtualisation is obviously changing the game for DR,” Lohmeyer said.

“Organisations should involve IT executives in the process of reevaluating their DR plans and then implement practices and solutions that ensure a successful return to full operations.

“We are concerned that organisations are not testing more frequently to improve their plans, and are not using adequate tools to reduce the overall business impact.”

Up to 44 percent of local businesses that use virtualisation have re-evaluated their DR plans, while 30 percent have not included their virtualised systems in their plans. Half of local businesses responding to the survey said they do not backup all of their virtualised systems.

Australian respondents blamed a lack of efficient technology (35 percent) and tight resources (35 percent)for the difficulty in backing up virtualised systems.

A fifth of local disaster recovery executions are triggered by system failures, according to the report, rather than testing. Respondents cited failures with hardware and software (20 percent); external security threats (12 percent); natural disasters (12 percent); IT problem management (10 percent); power outages (10 percent), and data leakage (8 percent).

About 34 percent of the DR tests that fail average a recovery time objective between 9.54 and 14.67 hours.

Executive buy-in from CIOs, CTOs and IT directors in DR has fallen by from 55 percent last year to 33 percent, according to the figures.

Australian respondents had more mission critical applications (64 percent) than the global average (56 percent) which Lohmeyer said increases the cost of availability. He said businesses should cut server sprawl and increase capacity, and review data centre configurations.

A profileration of tools for physical and virtual environments was blamed for raising training costs, reducing operation efficiency and isolating staff.

Lohmeyer said automated tools will reduce manual processing in DR while will help reduce downtime.