According to the Ponemon Institute “2016 National Survey on Data Center Outages”, power related issues are the leading cause of unplanned outages in data centers. And the total cost of downtime is $8,851/minute according to the same report.
All this comes with the growing realization that organizations rely on the data center to generate revenue. It’s no longer seen as a cost center or backroom operation.
While Data Center Administrators/Managers fully understand the cost of unplanned outages, they lack a clear idea of the kinds of preventative measures required.
Not surprisingly, one of the more popular responses is to throw more money or people at the problem. Many respondents to the survey felt that larger DC budgets and hiring additional staff was the way to go.
The majority of respondents took the approach of trying to fix the underlying cause of the problem. They recognized that investment in improved IT equipment or redundant IT equipment would go a long way to solving their outage problem.
Not just any IT equipment will do – they will require the right equipment to truly protect their data centers from downtime.
Most users and systems react to problems like power loss or power spikes once alerted by their monitoring systems. They still experience downtime but because the equipment is being monitored, they can quickly fix the problem and get back online.
A more effective approach is to choose a system that predicts potential failures and proactively mitigates problems such as power issues before they take the system offline. This level of intelligence is simply not available in traditional high availability or hyperconverged systems. Administrators and Managers need to closely examine traditional systems to see the gaps and look at intelligent platforms that bridge the divide to close these potential downtime causing gaps.
The question you should be asking is “are you really fixing the outage by simply reacting to it”?