California has a handy tool to help figure out when a car’s oil needs to be changed. Every 3,000 miles used to be the standard, but today’s vehicles can go as high as 15,000 miles depending on the model. Still, aside from fully electric vehicles, every car needs regular oil changes, just like every data center needs regular maintenance. What happens with improper or non-existent maintenance?
Both the car and the data center will not operate at optimum efficiency or, worse yet, will fail and cause unplanned downtime and business impact. The same can be said for replacing parts like brakes or tires. Without upkeep, equipment will fail and further damage the vehicle — just like delaying aged-asset replacement in a data center can have a ripple effect.
Creating a maintenance plan
Car manufacturers and industry organizations provide overall guidelines and standards for the base of maintenance care, but when it comes to data centers, managers must consider their facility’s specific design, equipment and operational requirements and customize a maintenance plan accordingly.
Developing site-specific standards starts with assessing three factors:
1. Business risk: This depends on the industry type (which aligns with the mission criticality of the data center to the overall functioning of the business), the technology being used, location and how closely data center operations are tied to business operations (i.e., colocation and cloud providers).
2. Inventory: Understand what needs to be protected and take note of the age make and model of the equipment.
3. Manufacturer’s recommendations: Know the manufacturers recommended maintenance schedule for processes like changing the generator oil or uninterruptible power supply (UPS) capacitors to help prevent high-cost repairs and extend the life of the device.
Data center standard design and implementation
Once these three elements are clearly outlined and reviewed, you can properly map out the standards that are right for your data center’s maintenance plan.
From a high level, this starts with understanding the core requirements and basic functions for each piece of equipment — regardless of the manufacturer. For instance, any UPS should provide emergency power to a load when the input power source or main power fails and hold that load until the generator and/or utility power kicks in.
Next, apply the policy standards to every site within the data center portfolio. This process is often overlooked, but it’s important that each site manager map out how the high-level standards apply to their particular facility in order to develop specific methods of procedure for their site.
Remember, there is no one-size-fits-all approach; even identical facilities can have varying standards based on their risk profile. A data center located in a hurricane-prone area, for example, might test emergency and backup power systems more frequently during storm season, than a similar facility located away from such natural disasters.
Whether changing oil in your car or batteries in your data centers, the key to longevity and health is in good planning and regular upkeep. Investing the time and resources proactively can ensure optimal performance. Read more about how to create a framework for developing and evaluating data center maintenance programs.
Editor’s Note: This blog was adapted from an article that originally appeared in Data Center Journal.