Optimizing a data center makes the facility increasingly attractive to clients, more agile when meeting needs, and less prone to downtime, among other benefits.
Here are six things data center professionals can do to make their facilities more efficient.
Resource management in a data center requires a multi-prong approach that begins with understanding what's causing unnecessary energy usage. For example, if servers are idle but still powered on, they're consuming energy that could be used in other ways. After identifying problem areas, data center operators may invest in equipment that senses power needs and makes automatic adjustments.
There are software-defined power interfaces that allocate energy in a data center to the components that need it, shifting as those requirements change. Some even have batteries that charge when not in use, providing a data center with backup power.
Some data center optimization practices don't require spending more money. Data center managers can review current internal procedures and consider whether some of them hamper efficiency. For example, it may be necessary to take a closer look at data center staffing and ensure that the balance of personnel working on particular shifts is appropriate for customer needs.
Changing the number of people who work a given data center shift could allow for a more appropriate balance of human resources. In addition, adjusting things that way could accommodate for shortages without hiring new employees and thereby spending more money.
Some resource adjustments do require financial expenditures. It's smart to examine client contracts and ensure that the facility is still able to meet the needs specified in service level agreements. When taking on new clients, facilities should assess risk and efficiency to gauge whether it's financially rewarding to accept those customers and give them the outcomes they expect.
According to an IBM study of global data centers, more than three-quarters of the overall respondents intended to recruit outside help when planning upcoming projects. As such, an optimized data center is one where its managers realize that shortcomings exist and let external parties fill them.
Data center optimization also relates to maintaining compliance. If a data center has not performed a compliance check recently, the news of an upcoming audit could make data center employees scramble to take action before that inspection occurs.
Data center audits evaluate physical assets, but those involved may also check to ensure certain types of data are stored in accordance with regulations. For example, there are specific regulations for storing health data. Also, the General Data Protection Regulation (GDPR) lets people living in European Union countries request that companies delete specific information collected about them.
Some of the ways to reduce data center audit headaches include having a wireless physical asset tracking system in place that monitors conditions and sends relevant information in real-time. Some intelligent gateways also perform automatic software updates.
In general, it's better to periodically ensure compliance before an audit happens. Then, if an inspection does occur, it won't become so hectic.
Keeping a data center within the proper temperature range is essential for trouble-free operations and prolonging the lifespan of the equipment. It's possible to have a significant impact on data center efficiency by looking for improved ways to cool data centers. Doing so might include enhancing the existing air-conditioning system or containing exhaust heat.
Google has worked on a multi-year project to create a cooling algorithm for its data centers that collected information about conditions, then gave recommendations to human operators. That approach caused an approximate 40-percent reduction in cooling costs. However, in the summer of 2018, Google let the algorithm run independently. Even so, people can intervene if it seems to do something too risky.
Even if a data center does not implement something as high-tech as artificial intelligence, its managers can still determine if the installed cooling systems are the best available or if more highly optimized options align with the facility's budget.
Some of the world's biggest data centers depend on renewable energy to operate efficiently. Hydroelectric power and wind energy are two of the most widely used options.
Indeed, a facility's size and financial resources may make it unfeasible to make renewable energy a primary and immediate concern when optimizing a data center.
However, when facility managers think about transitioning to renewable energy when possible, they should study the techniques used by hyperscale data centers and see which methods worked best in those cases. It's also useful to remain aware of local climate and weather patterns and how they could affect renewable energy availability.
In May 2018, speakers at a data center conference highlighted renewable energy's transition to a must-have amenity, whereas data center customers once considered it merely nice to have. Optimizing to take advantage of renewable energy could make a data center more future-proof while catering to sustainable-minded clients.
Consistency is one of the aims for an optimized data center, but humans could unintentionally interfere by leaving lights on or doors open as they move throughout the facility.
Unnecessary illumination wastes energy, and a door left ajar could cause temperature changes that make the climate control equipment work harder. Automatic sensors can help keep the data center's environment more stable, taking care of aspects that humans often overlook.
These tips offer valuable guidance for data center personnel intending to make their facilities more efficient.
However, it's necessary to remember that emphasizing data center efficiency for only a short time is not adequate.
Instead, operators should strive for efficiency in everything they do and adopt both short and long-term ideas of what optimization means for their organizations.
Kayla Matthews writes about data centers and big data for several industry publications, including The Data Center Journal, Data Center Frontier and insideBIGDATA. To read more posts from Kayla, you can follower her personal tech blog at ProductivityBytes.com.