Scheduled downtime: Minimise the pain by making it rare  

In the age of the ‘Always-On Business’, downtime is becoming increasingly unacceptable - even when it’s planned and announced in advance.

Any business that has online components must accept that its customers, employees and suppliers demand constant, uninterrupted availability. Even in the smallest hours of a Monday morning, someone is going to want to make a purchase, check an account or log into a system. There are occasions when there’s no alternative to taking an entire business offline for a few hours, but they should be rare. It is therefore a CIO’s job to deliver the availability that customers and other key stakeholders demand.

Further to this, thanks to the virtualisation advances of the past decade, always-on availability is not difficult to achieve. The modern data centre can and should be designed from the ground up with properly protected data and redundant systems, with guaranteed rapid recovery to get uptime as close to 100% as possible. Even in the worst-case disaster scenario, recovery time and point objectives should be no more than 15 minutes.

The longer a system is down for, the more it costs a business both in terms of turnover, and in customer expectations. According to Veeam’s most recent Data Center Availability Report, enterprises across the world have unplanned downtime an average of 13 times a year, totalling 51 hours per annum. Depending on whether the downtime affects mission-critical apps or not, that adds up over a year to an average cost of between $1.4m and $2.2m in lost revenue, as well as decreased productivity and missed opportunities.

In order to avoid continually losing this level of revenue, it is vital for businesses to achieve non-stop service and continuous protection, which has traditionally required a large investment in fully redundant systems with instant failover. Only the largest enterprises could afford these, and then only for their most important applications. But newer cloud, virtualisation, storage and other enabling technologies have made always-on capability far more widely accessible.

With these technologies becoming more affordable, enterprises can start using the right tools and begin to bridge the availability gap between the requirements of the always-on business and IT’s ability to effectively deliver business continuity and data centre availability.

It should be possible to restore a correctly backed-up virtual machine (VM) to the location of your choice in seconds. Enterprises can also create a replica environment that can continue to run core production systems so that there’s no downtime, even when a primary environment is being upgraded or undergoing maintenance.

It is recognised that downtime is a significant business risk that we need to protect against, but planned downtime should be equally unacceptable, because now it’s almost completely avoidable.