Average Cost of a Data Center Outage

More and more commerce and communication is happening through the web each day, and the importance of networks and data centers is higher than ever before.  So what can we do to make sure data center owners and operators aren’t losing money (and more importantly, creating unhappy customers)?  Well, let take a look at the root causes of these outages from the audited facilities:

UPS system failure (which includes batteries), cyber attacks and the dreaded “human error” account for 70% of the outages.  Most all of these outages were completely preventable and in many cases, the cost to prevent the problem was insignificant compared to the direct and indirect cost of the outage.

Generally speaking, here are some of the most basic tips to keep downtime from bringing you down:

1. Monitor UPS Batteries – Batteries are the weak link in the UPS system. Use remote battery monitoring to identify battery problems before they impact operations.
2. Use Intelligent Thermal controls with Cooling Units – These controls improve protection by monitoring component data points, providing unit-to-unit communications, matching airflow and capacity to room loads, automating self-healing routines, providing faster restarts and preventing hot/cold air mixing during low load conditions.
3. Perform Preventive Maintenance – An increase in the number of annual preventive maintenance visits correlates directly with an increase in UPS MTBF. Going from zero to one preventive maintenance visit a year creates a 10x improvement; going from zero to two visits a year creates a 23x improvement.
4. Strengthen Policies and Training – Make sure the EPO button is clearly labeled and shielded from accidental shut off. Document and communicate policies and conduct regular training.
5. Standardize and Automate Security Management – Use console servers to provide secure, remote access to servers to simplify patch management and provide early detection of attacks.

Cost of Downtime is a popular number and a useful metric to have when making the case for additional resources (human and equipment) to make sure your facility is always on, but it’s not the only metric that IT and Facility professionals should think about.

In the coming months, the Ponemon Institute will be releasing four additional reports as part of the Data Center Performance Benchmark Series, covering the issues of security, productivity, speed-of-deployment and cost-to-support compute capacity.

Submitted by Daniel Draper, Director of Marketing Programs for Emerson Network Power.