Firms greatly underestimate cloud downtime

The impact of a cloud outage has not been assessed by the majority of businesses that are moving across to the cloud, according to a new study.

The research by Veritas Technologies indicates that 62% of respondents across Australia and New Zealand (ANZ) have not fully evaluated the cost of a cloud outage to their business and are therefore ill-equipped to deal with the impact of an outage.

More than one-third of those surveyed expected less than 15 minutes of downtime per month, despite the fact that the average downtime due to cloud outages is 22 minutes per month.

While cloud service providers offer infrastructure-based service level objectives, the research indicates that many organisations fail to understand their own responsibility, in addition to that of the cloud service providers’, in ensuring that their critical business applications are adequately protected in the event of an outage.

The Truth in Cloud study, commissioned by Veritas and conducted by Vanson Bourne, surveyed 1200 global business and IT decision-makers. It revealed that almost all (99%) of the respondents globally reported that their organisations will move systems to the cloud in the next 12–24 months. In ANZ, close to a quarter (24%) also expect to outsource all on-premises infrastructure to the public cloud.

While migration to the cloud continues to accelerate, it is imperative that customers understand how an outage could impact their business.

More than half of respondents (51%) believe that dealing with cloud service interruptions is the primary responsibility of the cloud service provider. 71% of respondents also believe that their organisation’s cloud service provider is responsible for ensuring that their workloads and data in the cloud are protected against outages.

While cloud service providers have service level agreements in place, these are typically for the infrastructure layer and they hold the responsibility for restoring their infrastructure in the event of a cloud outage.

However, there are other key considerations customers should keep in mind that go beyond the actual infrastructure-level outage, such as bringing their applications back online, once the infrastructure is running again. Depending on the complexity of application interdependencies during restart and the amount of data lost during the outage, the actual time of application recovery may be far longer than the time of infrastructure recovery. An organisation may alternatively decide to be more proactive and failover applications back to their on-premises data centre or to another cloud. This would be the primary responsibility of the organisation, not the cloud service provider.

“Organisations are clearly lacking in understanding the anatomy of a cloud outage and that recovery is a joint responsibility between the cloud service provider and the business,” said Mike Palmer, executive vice president and chief product officer, Veritas.

“Immediate recovery from a cloud outage is absolutely within an organisation’s control and responsibility to perform if they take a proactive stance to application uptime in the cloud. Getting this right means less downtime, financial impact, loss of customers’ trust and damage to brand reputation.”

Image credit: ©castelberry/Dollar Photo Club

Follow us on Twitter and Facebook