Today, organizations from a wide variety of industries have their own data centers, e.g., telecommunications, government agencies, education, healthcare, and more. Regardless of your data center’s location or size, it is important to identify all areas that are ripe for improvement so that it can run more efficiently. Data center owners can consider streamlining operational procedures for on-site staff members, deploying advanced data center infrastructure management (DCIM) software, improving airflow management, enhance cooling systems, and more. These practices make good financial sense because they help reduce overall power and operating costs. With that in mind, below are four data center infrastructure management best practices you can put into action:
Track IT Load Demands
The goal of evaluating IT load demands is to determine if one has implemented the correct amount of uninterruptible power supply (UPS) and that these systems are running efficiently. Oftentimes, UPS systems are purchased according to the data center’s maximum IT capacity. In addition, UPSs end up being underutilized as the data center gradually fills up. Although modular UPS systems are available, they tend to be less efficient when underutilized. It is recommended that data center owners build out their centers in zones or stages, and only activate new appropriately sized UPSs as each stage is filled.
Leverage the Benefits of Economization
Economization strategies are designed to help reduce overall operating costs. Many organizations, however, tend to employ these strategies only when their cooling units are nearing end-of-life or during equipment upgrades/modifications. When it comes to economization, business owners do not need to purchase new equipment to take advantage of it. For example, plate and frame heat exchangers can be added to existing chilled water systems. After economization is in place, it is also important to enhance existing controls to automate the process so that more economization hours per year, especially at warmer temperatures.
Use Sensors to Monitor Cold Aisles
Research has shown that legacy data centers and colocation centers in the U.S. typically do not monitor cold aisle temperatures in a consistent manner. It is recommended that data center managers install sensors that can record temperatures so that trends on how hot return air and cold supply air is mixing can be established. This way, individuals can accurately identify areas that have potential efficiency gains. “Is an area overcooled?” “Does the area meet customer service level agreements (SLAs) for temperature?” These are some areas that can be better addressed with the right monitoring solutions in place.
Control Cooling Resources in Real-Time
The temperatures inside data centers need to be kept cool because of sensitive IT equipment. Some data center managers, however, make the mistake of running both redundant and required cooling units at the same time. They may have the impression that flooding the data center with cold air (giving the center two or three times the amount of cooling required) is the way to go. In order to achieve maximum efficiency, cooling resources have to be controlled in real-time. For example, one can install variable frequency drives (VFDs). These drives offer a best fit cooling solution where it cools the equipment down according to how hard it works.