The Art and Science of Cooling Data Centers

Rising energy costs, limits on resources and environmental regulations are a few of the challenges driving the need for energy-efficient sustainable data centers. Building a data center requires new focus, a holistic approach to your entire infrastructure, not just your IT equipment. Thermal analysis of high-density computing data centers drives facility design to optimize heating, cooling and server performance.

Engineers of server systems start with an assumption that they will have no static pressure difference from front to back. Servers take inlet air (supply air), use it to cool components and reject heat via outlet air (return air). The difference between supply and return temperature is referred to as ΔT and is a crucial design factor that affects all cooling systems.

Traditional cooling paradigms fight the airflow of servers; they overflow, or over-speed, or underflow the air supply. Overflow of supply reduces ΔT, reducing the efficiency of all cooling processes. Underflow of supply air requires servers to do more work, resulting in higher internal temperatures and fans working harder to pull air that they should not have to pull. Overspeed refers to situations in which velocity of air from CRACs is too high for a cabinet in the path to receive cool air, resulting in higher, not lower, temperatures.

KyotoCooling® changes many long-held preconceptions about cooling and represents the lowest risk option for a data center designer and owner. KyotoCooling® is about control … absolute control of the data center cooling system, irrespective of changing loads and environmental conditions, and without the need for operator intervention. KyotoCooling® is a total solution — tested and validated as a complete solution — designed with no single point of failure.

Begin with simplicity. To cool with air, you must control that air. Control means:

  • Not delivering more air than needed to cool with maximum efficiency
  • Cold air not being contaminated with hot exhaust air — recirculation or temperature contamination is responsible for inconsistent temperatures from floor to top of the rack

Typically, bypass and recirculation account for a 30 percent loss of efficiency in data center cooling.

We separate and control cold and warm air, resolving the first risk of cooling — control of temperature across long aisle spaces from floor to top of rack. The impact of this control is immediate and sensible — there is consistency in temperature and certainty in airflow delivered to the data center. There are no more anomalies and inconsistencies in supply temperature to The Art and Science of Cooling Data Centers 5 the data center, proven through empirical data, computational fluid dynamics and laws of engineering why this works.

Design the cooling process. We believe that it’s smart not to change the design conditions for optimal server performance. We precisely control the flow of air based on measured ΔT. We are able to maintain precise balance in pressure between the cold and hot spaces created by containment systems to deliver an optimal environment for server operating conditions. We do not underflow, nor overflow — we deliver a exact volume of air at a precise temperature to the cabinet face, even under changing load conditions, all without operator intervention. We provide exact cooling for the intended load. The amount of air required to unload an amount of heat can be determined by the permutated equation Q=m Cp ΔT. This means that for a given load in the data center, we can determine the fans, motors, compressors and coils required to achieve the desired result. These selections are based on certified performance curves of these subcomponents. These can be validated at a subcomponent level.

Cool the air. The patented design of KyotoCooling® uses a specially designed energy recovery device, known as the KyotoWheel™, for air-to-air heat transfer without bringing outside air into the data center. The KyotoWheel™ operates at extremely low rotational speed and has performed reliably in industry for 25 years or more under harsh operating conditions.

The KyotoCooling® units are deployed in a redundant N+1 up to N+N design. This assures continuous operation during maintenance, service or failure of a unit. KyotoCooling® is based on high availability design and operates continuously without outage for years at a time. In an N+1 or N+2 deployment, Uptime Institute Tier 3 and Tier 4 designs are achieved. The total KyotoCooling® solution has been verified by external sources and long-term tests over a wide range of ambient and load conditions. The sum of the parts of an integrated KyotoCooling® working solution can be independently tested and validated.

Control the complete cooling system. The sum is greater than the parts, and that philosophy is embodied and made true in our controls environment, the most sophisticated data room cooling control system in existence today. At no time can loss of any control or subcomponent remove the entire capacity of a KyotoCooling® system from the data center. The system is designed to act and react as necessary to achieve continuously stable operating conditions. Load changes are automatically stabilized to target predetermined operating metrics. This means the addition of load can be accomplished within minutes — irrespective of data floor location — automatically and without risk or disruption. No site level programming is required; the system is self tuning, autonomous across load, environmental changes and component adaptations.

We believe mission-critical cooling has to start with fact-based design. KyotoCooling® always states its design assumptions, delivers a solution to attain and manage supply control and containment, and uses the most conservative ΔT design to the client’s advantage. We are experts in controlling air and ΔT assessment, design and control. Mission-critical cooling starts with a solid understanding of the thermal environment on the floor and how to control air. That’s smart.