Peter Panfil is Vice President Global Power Sales, Emerson Network Power. With more than 30 years of experience in embedded controls and power, he leads global market and product development for Emerson's Liebert AC Power business.
Emerson's Liebert AC Power
I hate waiting in line and am as impatient as anyone I know - just ask my children. Many consumers venture out for Black Friday, but I prefer to do my shopping from the comfort of my keyboard. Over the past few years, more and more holiday shoppers are doing the same. More than ever before data center managers need to be ready for the rush of Cyber Monday. According to the Adobe Digital Index 2012 Online Shopping Forecast, Cyber Monday 2012 online revenue for the retail sector is expected to reach $2 billion, growing by 18 percent year-over-year. My question to you: Is your data center ready?
No One Likes Downtime
For companies that rely on data centers to deliver business-critical services, downtime has always been a key concern, especially during the holiday season. According to Ponemon Institute’s “National Survey on Data Center Outages,” 95 percent of companies have experienced an unplanned downtime event within the past two years. Types of outages cited total and partial data center outages, and device-level outages. However, even though more than 60 percent of these companies rely on their data center to generate revenue or support e-commerce activity, less than 35 percent believe their data center utilizes critical system design and redundancy best practices to maximize availability.
To gain a better understanding of which vulnerabilities contribute to such a high occurrence of costly downtime events, the survey asked more than 450 data center professionals to cite the root causes of the outages. The majority of data center downtime was related to inadequate investments in a high-availability infrastructure. In fact, more than half of data center professionals polled agreed the majority of downtime events could have been prevented. Below are three backup matters data center managers should assess to ensure the performance and efficiency of their facility.
1. UPS Battery Failure
While batteries may be the most “low-tech” components supporting today’s mission-critical data centers, battery failure remains the leading cause of unplanned downtime events. In fact, studies conducted by Emerson Network Power indicate battery-related failures account for nearly half of all uninterruptible power supply (UPS) system failures over the life of the equipment.
To safeguard backup power systems against unplanned or premature battery failures, data center professionals should take steps to ensure battery maintenance best practices are observed. In addition to standards outlined by the Institute of Electrical and Electronics Engineers, manufacturer schedules for maintenance checks should be followed to ensure batteries are properly maintained and/or replaced before they pose a risk for mission-critical applications.
While proactive battery monitoring and maintenance are critical to maximizing UPS availability, data center professionals should also keep charged spares on-site to cover any cells that may have expired between service visits.
2. Exceeding UPS Capacity
High-density configurations have become more common in recent years as data center managers seek to achieve increased throughputs from their existing infrastructures at the highest efficiencies possible. In fact, the average power draw for an individual rack can exceed 10 kW in some high-density data centers. With such high densities common during peak hours, the capacity of a single UPS system can be quickly exhausted, leaving critical IT equipment unprotected and the UPS system at risk for overload and failure.
If operating under a single-module UPS configuration, it is important to have a thorough understanding of the typical loads experienced throughout the data center. By measuring and trending the critical load via an integrated monitoring and management solution, data center professionals can gauge the typical power draw of their IT equipment over time to confirm whether their infrastructure is backed by enough UPS capacity.
Establishing a redundant UPS architecture also enables data center professionals to increase capacity of their backup power system, with the added benefit of eliminating single points of failure. Parallel (N+1) redundancy remains the most cost-effective option for high availability data centers and is well-suited for high density environments where future growth is certain. Alternative high availability systems like reserve and distributed reserve systems maintain availability while improving utilization and efficiency.
3. UPS Equipment Failure
In addition to exceeding capacity, general UPS equipment failure is another leading cause of downtime cited by the survey of data center professionals, with 49 percent reporting an equipment failure within the past two years. With this in mind, it is important to consider a number of key factors that can affect the reliability and overall lifespan of UPS equipment.
System topology is a significant consideration when evaluating UPS reliability. Generally speaking, two types of UPS are found in today’s data centers: line-interactive and double-conversion. While double-conversion UPS systems have emerged as an industry standard in recent years, many legacy and/or small data centers may still be using line-interactive systems in their critical infrastructures. Therefore, when evaluating each topology, data center professionals should consider the criticality of their data center operations.
Data center professionals should also consider the overall durability of their UPS system.
Some UPSs are designed with integrated fault tolerances for environmental and load conditions and a variety of redundant internal components; including fans, power supplies and communications cards. These features enhance reliability and enable the UPS to maintain availability between service visits even in the event of an internal component failure.
However, regardless of the type of UPS used, preventive maintenance is necessary to maximize the life of critical equipment and minimize the occurrence of unplanned outages. All electronics contain limited life components that need to be inspected frequently, and serviced and replaced periodically to prevent catastrophic system failures. If not serviced properly, the risk of unplanned UPS failure increases dramatically.
Good hunting, and don’t forget, I like bacon.
Industry Perspectives is a content channel at Data Center Knowledge highlighting thought leadership in the data center arena. See our guidelines and submission process for information on participating. View previously published Industry Perspectives in our Knowledge Library.