Free cooling can dramatically reduce the power needed to cool the data center. Yet very few (if any) colocation providers leverage it. A new approach to data center cooling means colocation customers get the energy efficiency and lower costs of free cooling with the reliability and locate-anywhere flexibility of a chiller plant.

Free cooling, which is the use of naturally cool air instead of mechanical refrigeration, can dramatically reduce the power needed to cool the data center. For colocation tenants, that means dramatically reduced costs.

But traditional free cooling systems aren’t practical for most colocation data centers. There are three primary reasons why:

1) In locations where it gets too hot and/or too humid, traditional free cooling doesn’t work – or it brings cold aisle temperatures to levels that most colocation tenants simply aren’t comfortable with.

2) Without conditioning (which can be expensive and reduces the efficiency of the system), outside air can introduce contaminants to the data center and/or make the data center too humid or too dry – all of which can cause an outage.

3) Traditional free cooling systems typically aren’t able to meet the cooling needs of high density IT environments.

So there are very few (if any) colocation providers leveraging free cooling at all – they still use the old chiller plant/forced air technology. Their customers, then, are trading energy efficiency and lower costs for location flexibility and reliability.

That’s a tradeoff that customers at Aligned Data Centers don’t have to make, thanks to our cooling technology (from the award-winning Inertech, also an Aligned company), which delivers ultra-efficiency and lower costs in any climate with Tier IV level reliability.

5 key differences between chiller plant/forced air systems and the Aligned Data Centers cooling system – and what they mean for you

1. They blow cold air in; we remove the heat

Our new approach to data center cooling was born out of the understanding that the data center cooling problem is actually a heat removal problem. So instead of blowing cold air into the data center, our technology removes the heat. A heat sink draws the hot air from the servers, passes it across coils and “neutralizes” the heat without chilling it – sending 75-77°F air to the server inlets at the front of the enclosure.

How it benefits our customers:
  • Removing heat at its source takes far less energy than making outside air cold and blowing it into the data center to mix with the hot air there.
  • Removing the heat is also more effective – we can take far more heat out than a traditional chiller plant can. That allows our customers to run higher densities (25 kW+ per rack) without having to spread their racks apart (which strands capacity).
  • The close-coupled heat removal system allows our customers to run racks at different densities in the same pod without worrying about hot spots. That’s not possible with a system that’s blowing cold air into the space.
  • Our system can flex according to demand; it’s dynamic, ramping up when IT loads are higher, and slowing down when they’re lower. In contrast, a traditional forced air/chiller system is static, which means it’s inefficient when IT loads are low, and can’t support above certain loads.
2. They push cold air up; we allow it to flow down

It’s a standard principle of thermodynamics: hot air rises. Our system takes advantage of that principle and absorbs heat at its source, allowing hot air to rise and cooler, denser air to settle where it is needed. In contrast, a traditional chiller plant has to push cold air up through the floor to where it’s needed and pull hot air down from the ceiling to get cooled – requiring more energy, and raised floor.

How it benefits our customers:
  • Our customers maximize every square foot of capacity of their IT footprints. In contrast, with a traditional chiller plant/forced air system, the server racks have to be positioned very specifically over those cold air vents, so their customers often have to take more space than they need.
  • Following the principles of thermodynamics and allowing cooler air to settle uses much less energy than forcing cold air to rise. Where in a chiller plant/forced air system fan power is about 10% of IT load, in our system it is 1%.
3. They use power- and water-guzzling chiller plants; we use a power- and water-efficient air-cooled adiabatic rejection system

Our air-cooled adiabatic rejection system comprised of two major components: a fluid cooler with an indirect evaporative cooling mode, and a compressor trim unit. The distinctive design combines both condenser heat rejection and water-side economization functions into the same product. In contrast to traditional chillers, this system relies on free cooling most of the time, even in hot climates. When temperatures can’t support 100% free cooling, the system makes use of indirect evaporative cooling. The cycle consistently and effectively manages water and compression power consumption.

How it benefits our customers:
  • Our customers get up to 80% savings on energy and up to 85% savings on water when compared to a traditional chiller system.
  • We guarantee an annualized PUE of 1.15 and use that in our pricing calculations – so you get the cost savings generated by our efficiency.
  • Because our system is so much more efficient than a traditional chiller, we don’t have to build as much infrastructure to support peak load. That means our customers get up to 60% more IT capacity out of the power utility.
4. Their cooling system has more than 1000 moving parts; ours has less than 10

When Inertech designed our cooling system they dramatically simplified the cycle and removed the complexity. Where a traditional chiller plant has more than 1,000 moving parts, our cooling system has less than 10. Furthermore, the system was architected for reliability – it runs active/active (2N by design). And the heat removal system uses refrigerant, so there’s no water in the data hall.

How it benefits our customers:
  • Less complexity and fewer moving parts means fewer failure points and thus higher reliability.
  • 2N by design means any part of the system could fail and it wouldn’t be a failure point for the data center. (That is very unusual in a mechanical system.)
  • No water in the data hall means no risk of leak. In a traditional chiller system, water valves are huge opportunities for failure.
  • All told, the cooling system is as reliable as a Tier IV chiller plant – 99.9999% reliability.
5. They build in multi-megawatt increments; we build in 500 kW increments

The smallest chiller plants are designed to support multi-megawatts of IT load. Procuring new chiller plants and then bringing them online takes many months. For those reasons, most colocation data centers build their chiller plants to support the maximum load of the data center. That means until the data center reaches full capacity, the chiller plants are running at less than full load, which is inefficient.

In contrast, our heat rejection system is built in 500 kW increments, which enables “right sized” deployment of cooling infrastructure in line with IT demand.

How it benefits our customers:
  • In addition to higher levels of efficiency (though our system is as efficient at 10% load as at 100% load) right-sized cooling infrastructure also enables capital cost deferment – which supports our consumption-based pricing model.

The benefit of the Inertech cooling system in use at our Phoenix data center and Plano data center is that our customers don’t have to trade energy efficiency and lower costs for location flexibility and reliability. They get ultra-efficiency and lower costs no matter the climate, with 99.9999% reliability.

Learn more about the technology behind Aligned Data Centers’ cooling system. Or see it for yourself – schedule a tour of the Phoenix colocation facility or Plano colocation facility.