Recent research concerning data center preventive maintenance, performance and downtime paints a positive picture of current trends. It also highlights why serious attention is still necessary in these areas.
Findings published by Uptime Institute show data centers across the globe experienced less downtime in 2021 than in 2020 – but when they did, the consequences were more complex and costlier to deal with than in the past. Among facilities that experienced a serious outage in 2021, 62% reported losses in excess of $100,000. That number was 56% in 2020.
The following steps outline the necessary features and considerations for creating a successful data center preventive maintenance strategy. It is the difference between courting losses and failure or thriving in a competitive industry.
1. Determine the Best Timing for You
It is essential to think through the logistics of “when,” “how often,” and “by whom” before you put any concrete plans into motion. Remember that maintenance should make your life easier, so any efforts to add or improve maintenance strategies should function harmoniously with your existing daily operations.
A sensible place to start is to determine your periods of peak demand and fashion your maintenance schedule around them. Preventive maintenance can disrupt business operations if it happens at an inopportune time.
2. Set Specific Goals
Another place to lay thoughtful groundwork for your data center preventive maintenance plan involves goal-setting. “My goal is maintaining my data center,” you think. But there is more to it than that. Here are some more detailed and advanced potential goals of data center maintenance:
- Maintain or improve facility efficiency
- Upgrade hardware and software
- Monitor and respond to cybersecurity threats
- Carry out general cleaning and maintenance
- Inspect the facility’s physical security
You know your data center better than anybody and will know where efforts would be best spent. Once you have outlined your maintenance goals in sufficient detail, the following step will help you determine and assign responsibility for the resulting action items.
3. Foster Subject-Matter Ownership
Reference the goals you compiled earlier and break them down into achievable tasks. Some of these tasks can easily be done in-house by your personnel. Furthermore, some of these action items could prevent issues from developing. Cultivate a sense of process ownership – and explain what is at stake – as you assign regular maintenance tasks such as:
- Visual inspections
- Cleaning
- Stress tests and backup tests
- Basic repairs and replacements
Other, more complex goals and tasks – like maintaining IT and energy efficiency, upgrading hardware and software or maintaining cybersecurity preparedness – may best be handled by a third-party maintenance partner.
4. Call on OEM Resources
Your business does not exist in a vacuum and your maintenance strategy should not, either. You should already have a wealth of information at your disposal from your original equipment manufacturer. These manuals and guidelines lay out:
- All necessary care routines.
- Recommended service intervals.
- Machine-specific best practices.
- General optimization suggestions.
- Recommended credentials for maintenance personnel.
OEM checklists for machine tending, cleaning and maintenance will maximize the useful lifetime of your equipment and ultimately save you money in the long run.
5. Keep an Eye on the Climate
In any context, the best way to prevent excessive maintenance tickets and downtime is to prevent problems from manifesting to begin with. A good data center preventive maintenance strategy should seek to take control of any variables within your reach. Start with climate.
It is becoming more common for data centers to become unresponsive in times of extreme heat. Step one in your preventive maintenance strategy should be to familiarize yourself with the latest and most comprehensive humidity and temperature guidelines. ASHRAE (American Society of Heating, Refrigerating and Air-Conditioning Engineers) released its most recent HVAC recommendations for data centers in 2019.
6. Use Remote Monitoring Technology
It can be challenging to be proactive about conditions within data centers. Large and small cloud and computing infrastructure providers have struggled recently with the warming climate. As many as 45% of U.S. data centers have reported difficulty staying online during extreme heat events. IoT remote sensing is the answer to the climate’s increasingly unpredictable behavior – and to improving condition monitoring within data centers in general.
The IoT provides a glimpse into the interior and exterior climate conditions, but it offers equally valuable telemetry concerning IT infrastructure. Smart IoT systems can throttle cooling systems up and down automatically and proactively shift data loads from one server or facility to another if it anticipates business-impacting events.
You should also note that prevention is not the final form of effective maintenance – proper predictive maintenance is the next step. The flow of data from remote sensors and IT system telemetry alerts decision-makers of probable failures before assets become responsive. This reduces the likelihood of downtime and makes maintenance schedules easier to plan out.
7. Stay Organized
Finally, give a shout out to staying organized. You would be surprised how often this lesson needs to be re-learned, even by generally conscientious people.
Staying organized begins with knowing what you have. Gather all available information about each piece of equipment, including:
- Serial number
- Manufacturer
- Model number
- Date of purchase
- Date of last service
Keep all this information ordered in a known but secure location. Your in-house personnel and third-party maintenance partners will both benefit from not having to hunt down this information themselves.
It is recommended that you make “date of last service” the bare minimum of what you collect when it comes to maintenance history. Knowing what has been done to each asset over time – including past troubleshooting – will give maintenance personnel a better idea of where to begin next time the machine needs attention.
Hone Your Data Center Preventive Maintenance Strategy
Combined with appropriate technologies, your maintenance plan will:
- Help you identify and solve inefficiencies.
- Improve the reliability of your assets.
- Keep downtime to an absolute minimum.
- Save money through more efficient, longer-lasting equipment.
- Safeguard your reputation through consistent reliability.
- Give you a competitive advantage in a crowded, in-demand industry.
Precisely what your plan looks like depends on your goals, scale and overall business environment – including its literal climate. No matter what, a data center preventive maintenance strategy will help you maintain your equipment as well as the considerable value it represents for you and your clients.
The post How to Create a Successful Data Center Preventive Maintenance Strategy appeared first on Datafloq.