You are here
WP#78 - An Introduction to Coolant Distribution Units
Liquid cooling technology, in various forms, has brought significant improvements in thermal capacity and efficiency to the heat problem with IT infrastructure. In recent years, with the rise of more powerful processors, the introduction of graphics processor units (GPUs), and greater emphasis on efficiency, liquid cooling has made significant inroads to the high-performance computing (HPC) and enterprise data center marketplace. With numerous companies providing novel types of liquid cooling solutions, it has become advantageous for readers to compare methods of IT infrastructure cooling for performance and cost factors.
The main purpose of a Coolant Distribution Unit (CDU) is to supply coolant in stable conditions (flow rate and temperature) under changing operational loads. Typically, a CDU is required to enable a liquid cooling deployment which will allow users to deploy the most powerful computer equipment available at very high rack densities. Additionally, there is the benefit of lower operating expenditure costs (OpEx), which allow the equipment to pay for itself within a relatively short period of time. With fewer traditional air handling units needed in a data center to handle heat loads, along with a reduction in ducting and plumbing equipment, the upfront cost of a liquid cooling system deployment can provide capital expenditure (CapEx) savings for a reader.
While the economic considerations of liquid cooling make sense, there is still a practical argument to be won. Readers continue to be concerned with the risks of a liquid cooling deployment. There is a perception that liquid cooling can endanger IT equipment in ways that air cooling cannot. By constantly exposing the HPC and enterprise data center industries to the merits and benefits of liquid cooling technology, these points of contention can be overcome.