Data center: How to Solve server refrigeration faults? (1)

Network infrastructure design is a complex process. Every small decision may affect the servers in the data center. What is most likely to be ignored is the server cooling requirements, either misjudgment or serious underestimation, such errors occur from time to time in various environments. Historical experience shows that the ever-tightening IT budget forces data centers to ignore critical cooling and temperature control needs, even if IT is sometimes the case for large data centers. Only by effectively preventing server cooling failures can the applications running on the server never go offline, and enterprises can truly save money. So how should we do a good job in server cooling and how should we handle faults? Recommended by BKJIA: Russian abnormal madman overclock AMD to 7G liquid nitrogen to ensure refrigeration)

Server refrigerating capacity

Before handling faults in server refrigeration infrastructure, you must understand how the refrigeration system works and how much refrigerating capacity is required. refrigeration is closely related to the number of electrical equipment placed in the data center, the expected power consumption can be calculated based on the number of AMPS on the device. The larger the number of AMPS, the more refrigerating capacity is required. The simplest calculation method is to know the power consumption of each server or all servers). Generally, a single-CPU server requires 1 amps or 120 watts and 1 amps x120 volts = 120 watts ), dual-CPU, such as Xeon or AMD) servers require 2 amps or 240 Watts, and multiplied by the refrigerating capacity required for each Watt by about 3.4 BTU) to calculate the refrigerating capacity required by the server.

For example, a 120-watt server requires a refrigerating capacity of 408 BTU of 120x3.4). This method is used to calculate the refrigerating capacity required by all servers, ask the certified HVAC consultant about whether the calculation result is feasible.