White Paper explains how row-based cooling works

Popular misconceptions about row-based data-centre cooling are addressed in White Paper 208 from Schneider Electric, which explains that capture and removal of hot air, rather than the supply of cold air, is its essential purpose.

  • 9 years ago Posted in

The basis of row-based data-centre cooling is often thought to be the supply of cold air to cool IT equipment as it heats up while in operation. However, a more accurate and useful description is that it is a “hot-air capture” architecture that neutralises air heated by IT equipment before it has a chance to mix with the surrounding air in the room.

To maximise the effectiveness of row-based cooling, it is necessary to be aware of the design considerations and to clarify any misconceptions regarding the technique. This is the approach taken in a new white paper from Schneider Electric, a global specialist in energy management and automation, which discusses and refutes many of the common misunderstandings that have become commonplace.

Co-written by Paul Lin and Victor Avelar, both Senior Research Analysts with Schneider Electric’s Data Center Science Center, White Paper 208 “How Row-based Data Center Cooling Works”, describes in detail three of the most common misconceptions and explains how a “hot-air capture” architecture works in practice.

The goal is to remove the heat added to the air by IT equipment. Placing racks and cooling units in a pod allows the hot air to be captured by the cooling units so that there is no net heating of the room. Some techniques that can help achieve 100% hot air capture include the use of blanking panels and brush strips; the use of side air-distribution units; the use of deeper IT racks to facilitate air flow; and the use of air containment to prevent hot exhaust air from recirculating back into IT equipment inlets.




Three common misconceptions are addressed and refuted, namely that row coolers require turning vanes to direct cold air to the front of the racks, that row coolers are required in every row, and that row coolers cannot cool loads outside the pods in which they are located.

Given a well thought out design, turning vanes are not required at all. Using such vanes increases the capital cost of the equipment and takes up valuable space in a data centre. Their removal from consideration has obvious cost advantages.

Similarly, with a design aimed at achieving a high degree of hot air capture, it doesn’t matter in which row the coolers are positioned. It is possible for row coolers in a single row of a two-row hot-aisle configuration to cool racks in both rows.

Another way of expressing the third misconception is to say that row coolers can’t cool a large room or cool standalone equipment outside of the racks in which they are located. In fact, as variable cooling capacity devices they can sense if the overall temperature of a room has increased due to the addition of equipment outside their racks and respond by increasing cooling capacity to neutralise it.