Data center operators face constant pressure to save time, save money, and enhance efficiency — all while keeping things up and running. It’s not easy, especially considering the expanding scope and magnitude of power outages across the country. Eaton’s Blackout Tracker logged 3,879 separate U.S. power outages in 2016, the greatest number since the company first began tracking in 2008. Furthermore, the number of people affected by power failures nearly doubled from 2016 to 2017, with more than 36.7 million impacted in 2017 alone.
Unexpected outages not only threaten system efficiency, but lead to large unexpected costs associated with downtime. With another hurricane season close to wrapping up and an ever-escalating list of threats that leave mission-critical environments even more vulnerable to downtime, it’s important for organizations to be prepared with the necessary solutions and support to ensure uptime in the case of an event.
In this article, we’ll explore more on the dangers of downtime and provide tips for data center operators to consider when implementing a strategic disaster preparedness plan.
THE COST OF DOWNTIME
The obstacles that organizations face in protecting their systems against power issues are exacerbated by the fact that modern IT equipment is even more sensitive to electrical disturbances than it was previously. At the same time, ensuring uptime is more important to the critical functions of many businesses than ever before. As a result, power quality problems have become increasingly costly and complex, resulting in substantial losses for the companies affected.
Unexpected downtime can result in an interruption to operations, lost or corrupted files, hardware malfunctions, the inability to access the critical systems, and more. When considering the fortitude of power protection solutions, businesses must ask themselves an important question: Can we afford to lose $100,000 per hour? That’s the most recent price tag associated with just 60 minutes of downtime, according to a recent ITIC study.
Keep in mind those numbers above only represent an average. For many organizations, the costs are significantly higher. As reported by Business Insider, the one hour of downtime Amazon experienced on Prime Day in 2018 may have cost it up to $100 million in lost sales. It’s also important to remember that downtime isn’t only about dollars; a negative experience can significantly damage an organization’s reputation and cause customers to flee to competitors. Data and monetary losses from unplanned outages can even cause a company to go out of business.
While the industry downtime average is dependent on many factors, and monetary losses vary based on a broad range of elements — including revenue, business type, outage duration, the number of people affected, and the time of day an outage strikes — it’s safe to say that downtime is a demon to be avoided at all costs.
NINE WAYS TO PREPARE FOR THE UNEXPECTED
Despite the risk of downtime that weather and other unforeseen emergencies can pose, there are steps that organizations can take to prevent their organization from getting burned. For data center and IT managers looking to take a proactive approach, outlined below are nine steps to consider in ensuring systems are always protected:
Assess your overall power protection solution. If your facility isn’t already protected, determine the type of uninterruptible power system (UPS) you need, the best deployment strategy, and how much capacity is required to support your business. For organizations with UPSs already in place, it’s a perfect time to reevaluate your overall environment and ensure the existing UPS solution meets your current needs.
Deploy power management software. Power monitoring and management solutions significantly enhance UPS functionality, enabling IT personnel to remotely monitor a single UPS up to an entire enterprise network with an array of devices and components from different manufacturers. Even more, these solutions offer a broad spectrum of advantages that contribute to optimized power, enhanced efficiency, improved data protection, and lower overall costs. With adequate backup runtime, modern software can also facilitate the safe movement of data to a backup site to maintain business continuity.
Additionally, predictive analytics services have emerged as a way to shift power monitoring from a reactive to a proactive model. By collecting and analyzing data from connected power infrastructure devices, these services allow operators to address failure of critical power components before it occurs.
Implement remote management. Remote power management offers organizations a wealth of advantages, enabling administrators to oversee their entire network environment from any location in the world. Remote solutions are also used by many businesses to manage servers at offices and branch locations where there are no IT staff members. From powering cycle servers and devices to troubleshooting problems, remote management eliminates unnecessary service trips to isolated locations. By allowing administrators to securely control the power to every piece of equipment in the data center, remote management facilitates more efficient server management and reduces equipment downtime by giving administrators direct access to power control.
Don’t forget the PDUs. A power distribution unit (PDU) can be fitted with multiple outputs designed to distribute electric power, especially to racks of computers and networking equipment located within a data center. These devices help to provide reliable data center power distribution for both raised and non-raised floor applications. In addition, PDUs deliver effective power management and monitoring when coupled with an energy management system, which optimizes both utilization and availability down to the branch circuit level. With a solution that offers an interactive web-enabled interface, users can benefit from real-time monitoring for power quality status and fast response to power-related events.
Schedule a checkup for your UPS. Your UPS is the first line of defense against the many hazards, so make sure it’s in optimal health. One of the best ways to protect your investment is by engaging in a service contract that includes regularly scheduled preventive maintenance (PM). These visits can help detect a wide range of ailments before they become serious and costly issues. During a PM service call, a trained technician will inspect your system, test the batteries, and examine key components such as fans, capacitors, and internal connections. If you have a generator, test it regularly to ensure you have emergency power in the event an outage occurs.
Review your disaster recovery plan. Every change that has taken place within your organization since you first compiled your disaster recovery strategy — from adjustments in employees to new hardware and software implementations — makes your plan that much less effective. Be sure to update it on a regular basis. Not sure where to start? Take a look at some resources that can help you plan ahead to avoid downtime
Always be prepared. Never lose sight of the fact that power problems are equal-opportunity threats; in addition to seasonal weather, downtime can result from equipment failures, human error, copper thieves, and even obscure hazards such as drones, curious squirrels, and wayward snakes. Anything that disrupts power has the potential to bring your business to a halt. Make sure you have a comprehensive power protection solution in place to keep operations flowing in the event of unforeseen circumstances.
Choose your site carefully. When selecting a spot for your data center, it’s wise to consider the site’s ability to withstand severe weather and disasters, as opposed to simply choosing the facility with the lowest cost. Avoid a single point-of-failure by maintaining multiple active sites in diverse locations (preparing for a local or regional disaster). Organizations that cannot afford downtime should not store data in only one location — or even in two locations susceptible to the same disaster. Also keep in mind room selection for your IT equipment. Seek higher (centralized) ground whenever possible to avoid flood damage. Variety is key, and when the time strikes, you will be happy all your eggs aren’t in just one basement-level basket.
Communicate with your customers. If your organization does fall prey to downtime, customers will be more forgiving if you communicate with them. While 100% uptime is expected daily, it’s human nature to forgive, especially businesses that are transparent.
THE BIG PICTURE
There’s no denying how important it is for businesses to adequately protect against downtime. Unplanned outages not only threaten system efficiency, but can lead to large unexpected costs and headaches for staff. Whether it’s the storms that Mother Nature might pose during hurricane season, or other emergencies that can arise throughout the year, organizations must consider power as a critical element in a disaster preparedness plan.
Advancements in power management have created opportunities for organizations to ensure business continuity. Solutions like UPSs, power management software, and other power quality devices are strategically designed to help organizations avoid the threat of downtime disaster. By leveraging these advanced solutions in alignment with one another, data center and IT professionals can rest easier knowing they’re prepared for whatever comes their way.