Data Center Journal


Issue link:

Contents of this Issue


Page 17 of 32

THE DATA CENTER JOURNAL | 15 tHe proper tools w e've probably all heard the adage, "If a tree falls in the forest and no one is there to hear it, does it make a sound?" Data center airflow is similar—only differ- ent. To understand the path air takes in a data center, it helps to have the proper tool for the job. e correct tool in this case is computational- fluid-dynamics (CFD) soware. e purpose of a CFD package is to give users a color-coded visual repre- sentation of the airflow in your data center—blue and green good; orange and red, not so much—that's based on the proposed layout. As we shall see, not all raised-floor (or slab) layouts are created equally. Interestingly, estimates indicate that 80% of data center customers use CFD tools to optimize the original layout of the facility, but usage drops off precipitously aerward. is decision is logical, of course, if an operator never expects the layout of its data center to change. Unfortunately, it's rarely the case in practice. At a minimum, most data centers perform hardware refreshes every 3–5 years. And in a world where not having or not planning to have a cloud presence makes you the data center equivalent of Neanderthal man, maximizing the efficiency of your cooling infrastruc- ture is a basic operational requirement. cfd Modeling: essentiAl to dAtA center plAnning And operAtion As self-evident as the answer to the question in the title may seem, it never hurts to understand the underly- ing rationale for the procurement and use of CFD soware, if for no other reason than someone in the finance department may ask. Now more than ever, data centers are evolving entities in which it's easy to make deployments but less easy to undo them. A failure to plan for the potential consequences of revised data center layouts on airflow and cooling can, and usually does, deteriorate your data center's cooling capacity. In some cases, you may begin experience cooling issues at levels below design capacity. e result is a large volume of unused capacity— a phenomenon commonly called "stranding" capacity. As previously stated, an estimated 80% of data center operators—ei- ther themselves or, more commonly, through their data center provider— employ CFD modelling to determine the optimal layout of their new facility. If these efforts are in concert with the commissioning process, thereby removing extraneous variables that can affect future reporting values, the data center can be calibrated through the establishment of initial benchmarks to identify the performance standards that are unique to the site. In electing not to use a CFD tool in a post-turnover environment, operators forfeit the ability to evalu- ate the effectiveness of their cooling efforts and to evaluate the impact of a potential new configuration on the facility's airflow. In practical terms, this situation means operators can "pre- test" alternatives before applying the concept of cooling-path management (CPM) to ensure that they implement the most efficient solution. CPM is the process of step- ping through the full route taken by the cooling air and systematically minimizing or eliminating potential breakdowns. e goal is meeting the air-intake requirement for each unit of IT equipment. Cooling paths are influenced by a number of variables including: the room configuration, the IT equipment and how it's arranged, and any changes to the facility, such as air-handling-unit (AHU) settings, cabinet arrangement and equipment placement. To actively avoid cooling problems and inefficiencies that may creep in over time, CPM is therefore essential to the initial design of the room and to configuration manage- ment of the data center throughout its life. e establishment of bench- marks and the evaluation of cooling paths enable CFD users to institute a program of "continuous modeling." e value of continuous modeling is, as referenced above, the ability to test potential changes before moving the IT equipment. Numerous "what-ifs" can be addressed (and costs avoided) while providing all essential capabili- ties, such as availability, capacity and efficiency. Examples of continuous- modeling applications include the following: 1. Creating custom cabinet layouts to predict the impact of various configurations 2. Increasing cabinet power density or modeling custom cabinets 3. Modeling hot-aisle/cold-aisle containment 4. Changing the control systems that regulate VFDs to move capacity where needed 5. Increasing the air temperature safely and without breaking a temperature SLA 6. Investigating upcoming AHU maintenance or AHU failures in ways that can't be done in a production environment In each of these applications, the appropriate modeling tools work in concert with initial calibration data to determine the best method of imple- menting a desired change. e ability to actively identify the deviation from

Articles in this issue

Links on this page

Archives of this issue

view archives of Data Center Journal - VOLUME 51 | AUGUST 2017