Better cooling is definitely one reason why data centre operators are attracted to colder climates like Iceland, but there is a hole in this argument. Whilst free cooling is very efficient, in most mainland European climates, adiabatic cooling works really well and allows you to achieve almost the same efficiency.
In mainland Europe and UK it never gets hot and humid at the same time. A hot summer’s day in the UK, 35˚C for example, and the wetbulb temperature never actually gets above about 23˚C. Even accounting for global warming, if you ramped the wetbulb temperature up to 25˚C it really is the maximum humidity you need to account for.
Adiabatic cooling takes advantage of this by spraying water onto the dry air coolers located on the roof, you normally get about a 10C temperature variation, enabling you to maintain a temperature of 35C even on the hottest summer day.
We’ve designed, built and operate a data centre that makes use of that theory, also recommended under the ASHRAE guidelines, and we have no back up DX cooling except for our plant room. The reason is that we know the maximum wetbulb temperature will never go above 25˚C and modern servers are warranted up to 45˚C – therefore we have 10 degrees of edge room.
By letting the data centre run hot, with the occasional excursion up to 35˚C on a few days of the year, comes significant cost savings.
Memset’s data centre was built to the old CESG IL3 standards and they signed off on the resilience and architecture based on our designs and recommendations at the time. Therefore we expect our data centre to operate at a PUE of roughly 1.2 given that even the best free cooled data centre is unlikely to manage better at 1.1 PEU, we think it’s worth the extra 10 per cent on our power bill. In addition, being located in Dunsfold Park in Surrey, which houses a large scale solar farm, means our data centre is partially solar powered, which more than makes up for the 10 per cent difference.
For most services, cloud based or otherwise, latency is a big factor. Ideally, you want your data centre to be close to your customers and ideally in the same continent. That’s why you don’t tend to find US based data centres serving content to UK businesses or customers. Amazon’s Ireland based data centres are an exception. Some services are more latency sensitive, but having your data centre located in urban areas, ie. within 100 miles of London, the latency would be approximately 20 milliseconds.
When you start having to use undersea links or going to remote parts, like colder countries like Iceland for example, you definitely have to factor in a big latency difference. Even in a website, a human can see a delay of 50 milliseconds, therefore they’d be getting a delay of at least that for every single query. There are also questions to be raised about moving away from urban areas. Out here in Surrey we get about a 1.5 millisecond latency back to central London hubs, so for all services it’s sufficient and we’re outside the M25.
Risking power supply
The advantage of being in relatively built up areas is the mature and resilient power structure you have access to. It is a gross generalisation, but common sense that the further you go from major populated centres and the further north you go the sparser the population is, the worse your connectivity is. This is slightly mitigated by the fact that back up power is good. But any data centre operator knows you don’t want to be using it unless you have to. If you are having to deal with a brownout every two months you are elevating the risk factor. Whereas in urban areas it’s more likely you’d face a brown-out every two years.
In summer, you don’t need to go somewhere cold. You just need to let your data centre run really hot and sprinkle some water on the roof. That way you can avoid the potential risk around your power supply and you don’t have to compromise your latency.