Swedish Data Center Saves $1M a Year Using Seawater For Cooling 78
alphadogg writes "A data center in Sweden has cut its energy bills by a million dollars a year using seawater to cool its servers, though jellyfish are an occasional hazard. Interxion, a collocation company in the Netherlands that rents data center space in 11 countries, uses water pumped from the Baltic Sea to cool the IT equipment at its facilities in Stockholm. The energy used to cool IT equipment is one of the costliest areas of running a data center. Companies have traditionally used big, mechanical chillers, but some are turning to outside air and evaporative techniques as lower-cost alternatives."
Heh, they havn't met the Zebra musels from hell .. (Score:1)
That'll shut them down eventually.
Re: (Score:1)
Re:Heh, they havn't met the Zebra musels from hell (Score:4, Interesting)
The zebra mussel is a freshwater creature. I'd be surprised if they had large problems with these in their seawater cooling circuit. I guess the salt will get them first.
Re: (Score:2)
Apparently they could survive in seawater. [helcom.fi]
Re: (Score:2)
Apparently they could survive in seawater. [helcom.fi]
From your linked article:
The Zebra mussel has become a part of the Baltic coastal ecosystem in many areas around the Baltic Sea, but the distribution is patchy, partly depending on the availability of suitable habitats and limited to areas of less saline water.
Re: (Score:2)
Stockholm's "sea water" is what most people would describe as brackish - the salt content is much lower than larger bodies of water. While I suspect that the Zebra mussels would still find it too salty to thrive, you might find that individuals can adapt and propagate.
Re: (Score:2)
As others have mentioned, this is seawater. But even if it were freshwater, power plants and municipal water supplies draw from rivers and lakes infested with zebra muscles. They are a PITA, but they work around them.
OK (Score:1)
strange....just $1 million? (Score:4, Informative)
So...from the article:
So "today" per 1MW of IT load, it would cost $5.4million / 5.5MW or $981818.18 ( 54/55 million $ per MW or 0.981818182 x million $ per MW)
$2.6 million - $0.98 million > $1 million
Now, if he wanted to cool 5.5MW of IT load, it would cost him $14.3 million with the old method vs $5.4 million with the seawater method.
Even if you account for the cost of the third-party...$14.3 million vs $5.4 million is a big difference.
Re:strange....just $1 million? (Score:4, Interesting)
Strangest thing is - Sweden is a relatively cold country where people pay for heating. And for hot water in the summertime. Can't all this excess (heat)energy be put to good use instead of dumping it to the sea?
Re: (Score:3, Informative)
Re:strange....just $1 million? (Score:4, Interesting)
Some rural industrial estates were using their hot air from their cooling systems to grow plants.
One placed I worked in had the external parts of their air conditioning in a ground level sheltered car park. The heat was so incredible, that you could comfortably walk around in this bubble of warm air in a T-shirt or short-sleeve in the middle of Winter. The only was homeless people wandering by and building makeshift tents around one or more of the units in winter, tripping various CPU temperature alarms.
Re: (Score:3, Informative)
Not easilly.
Computers typically use air cooling, the exhaust temperature of a computer is not very far above the intake temperature and the intake temperature is typically arround normal room temperature or lower. So the exhaust temperature is likely to be barely above normal room temperature making moving the heat arround difficult.
You could raise the intake temperature to the computers but doing so would have significant disadvantages. Firstly it would reduce the ammount of time you had between cooling eq
Re: (Score:1)
Re: (Score:2)
Warm the water directly (Score:2, Insightful)
So instead of warming the atmosphere, they're warming the seawater.
Cuts to the chase neatly that. Let's all of us dump our excess heat into the ocean and see how if works out better in the long haul.
No free lunch.
Re:Warm the water directly (Score:5, Informative)
Let's all of us dump our excess heat into the ocean and see how if works out better in the long haul.
The article said the warm water is sent to heat pumps to warm up houses in the town. They don't say if they are able to bring the temperature back down to the original levels or not, or even if the water is pumped back into the ocean.
Re:Warm the water directly (Score:5, Interesting)
Most thermal systems, be it in cogeneration/district heating, or even traditional power stations, still end up dumping some residual heat as waste into the environment.
It seems nuts, but it gets to a point where the temperature differential/gradient is simply not enough to justify an industrial process to recover the heat efficiently.
For example, if you were trying to heat your house with water that was only a few degrees above ambient, well, you'd probably not be very happy.
Still, sometimes it works out OK, like the example (in France, from memory), where waste heat from a nuclear reactor is used to heat ponds to grow tropical shrimps, and greenhouses for fruit.
By the time the water finally returns to the river, thermal impact is virtually zero, minimising local ecological disruption.
Re:Warm the water directly (Score:5, Informative)
Read the article: after leaving the data center, the heat is sent to a heat pump where it's used to heat houses.
Re:Warm the water directly (Score:5, Funny)
Read the article
Quit trolling.
Re: (Score:2)
Stop making sense.
Re: (Score:2)
after leaving the data center, the heat is sent to a heat pump where it's used to heat houses.
which is dumping it back into the environment.
Re: (Score:2)
If you're going to be pedantic, I've got you out-gunned.
Re: (Score:2)
So instead of dispersing the heat in the air directly (plus the heat generated by the chillers) It is used to heat houses. Even if the warmed water was dumped directly into the sea it would make little difference in the temperature of the water.
Re: (Score:2)
Human generated heat actually warms the environment closer to the ground which exacerbates the need to run air conditioning.
Overall, it's safe to say all mammals are contributing heat to the planet, but humans are doing a much more extreme job of it..
Re: (Score:2)
Recent articles point to Denmark (ca next to sweden) as having issues that fishermen are getting swordfish and other tropical species in their nets as by-catch, while the normal (local) fishies are moving north - all due to shift in the water temperature.
Re: (Score:2)
Pirate Bay (Score:1)
Reading between the lines of TFA (Score:2)
It doesn't pump seawater directly through its cooling systems. Instead, the seawater goes to a heat exchanger where it's used to cool fresh water. It has to clean the exchanger fairly regularly, but Coors said it's a simple maintenance job.
The kind of job any bartender [slashdot.org] will be soon very happy to do for the dole.
But it had to run the chillers only a few hours last year, Coors said, when the government ordered its partner to stop pumping seawater. Coors isn't certain why that was, but he believes it's for environmental reasons. "I think it's to protect the jellyfish," he said.
In Norway:
1. the govt has a partner
2. the jelly fish have govt protection only a few hours each year.
(note: Poe's law suggest a smiley. Here it is: *big-grin* )
Re: (Score:2)
In Norway:
1. the govt has a partner
2. the jelly fish have govt protection only a few hours each year.
Exactly how did you deduce that from an article about a data center in Sweden ?
Re: (Score:2)
Re: (Score:2)
Seawater is nasty (Score:2)
I hope they hired a marine engineer to work out the anti-fouling issues. The system may work great now, but in a couple months every single surface exposed to seawater will be covered with barnacles and algae. The article mentions cleaning heat exchangers as part of maintenance, but some of this crap can't be scrubbed off without a chisel.
Re: (Score:2)
Re: (Score:2)
I've been to Australia and the water in the Pacific is nothing compared to the water in Baltic Sea. It is a green putrid mess compared to the clear ocean water. Baltic Sea has so low salt concentration that all kinds of green stuff and weed live in it really well. All the nitrogen from farm lands also plays in to this.
So, the problems in Australia and Sweden might be quite different.
Re: (Score:2)
Or just any mechanical engineer (Score:2)
Re: (Score:2)
I hope they hired a marine engineer to work out the anti-fouling issues. The system may work great now, but in a couple months every single surface exposed to seawater will be covered with barnacles and algae.
The OTEC guys say that biofouling only occurs in their heat exchangers that are exposed to surface water. I don't know if that is because of the temperature of the water or because of a lack of organisms in the deeper water. But since these data center guys only care about cold water which they can get from the deep, they don't have to deal with the most of the problems typically associated with sea water.
http://www.otecnews.org/portal/otec-articles/ocean-thermal-energy-conversion-otec-by-l-a-vega-ph-d/#e [otecnews.org]
Re: (Score:2)
There is no deep water in the Baltic Sea: it's only about 50 meters deep in most places. The OTEC guys are talking about depths 20 times that.
Re: (Score:2)
50 meters? That's not a sea, that's practically a shoal.
Re: (Score:2)
The system may work great now, but in a couple months every single surface exposed to seawater will be covered with barnacles and algae. The article mentions cleaning heat exchangers as part of maintenance, but some of this crap can't be scrubbed off without a chisel.
This is a problem which has had a worked solution for at least 50 years. The largest and one of the oldest running oil refineries (built in the 50s) in Australia has used seawater for cooling since the beginning. They have orders of magnitude more heat exchangers and larger heat loads, not to mention complicated metallurgical requirements to boot. In comparison using a water to closed loop coolant system is a walk in the park engineering wise.
Back then they didn't do heat recovery though minimal effort is m
Using the cold environment isn't terribly new. (Score:5, Interesting)
Back when our basement data center housed 70s and 80s era IBM mainframes and their accoutrement (a dozen or so tape drives and a huge 3380 farm) , the building vented cold upstate NY winter air into the DC.
A few years after the final ECL mainframes and 3380s were replaced by "z" mainframes and EMC SANs, the vent was blocked up.
Re: (Score:2)
AC units for data centers can be ordered with economizers, which will vent outside air in when there's enough cooling potential. They will filter it, etc.
Interestingly enough, these units have to be freeze-rated also.
Location location location (Score:2)
Re: (Score:2)
I am routinely a bit confused as to why datacenters aren't predominantly located in places with colder climates. Free cooling from the outside during the winter and whatnot. Is there simply a lack of infrastructure to make an ultrahigh-bandwidth line out to...say...northern Montana?
Basically, yeah.
A lot of the networks have expanded where there's people: early networks grew up around universities and government facilities (often located in or near major population centers), companies later grew up (or migrated to) where the tech people were, and things more or less grew organically from there.
Take a look at Level3 [level3.com]'s network map for the US: there's a lot of facilities in areas where there's a lot of population: SF Bay Area, Los Angeles, New England, etc. Florida has more than I would o
Re: (Score:2)
This has got to be more efficient (Score:4, Insightful)
Nice to see a plan which is a win for the environment on top of being a money saver.
Re: (Score:2)
The power plant side of the equation is unchanged: the energy it creates depends on there being a temperature gradient.
Nothing special (Score:4, Informative)
This isn't exactly unique or special. [wikipedia.org] Most of downtown Toronto is covered by the cooling grid from one such deep-water lake cooling systems, and I know of at least one datacenter (one of if not the most critical in the country) that uses the service.
It's not that salty (Score:5, Informative)
So does Google in their datacenter in Finland (Score:3, Informative)
http://www.google.com/about/datacenters/inside/locations/hamina/ [google.com]
Here is a video about Googles sea water cooling system:
http://www.youtube.com/watch?v=VChOEvKicQQ [youtube.com]
Re: (Score:3)
If you'd bothered to read the article instead of trying to look like a wise-ass on the basis of ten seconds of reading, you'd know what they do with the waste heat.
Re: (Score:2)
I cool my cottage similarly ... (Score:5, Interesting)
I'm not using sea-water so maybe this is only tangentially interesting ... The water that comes out of my 10gpm well is at 8C. When I had my new forced-air furnace installed, I asked the installer to put in an evaporator coil to prepare for future air-conditioning... Cost me an extra $180. Later I removed the orifice, hooked up a solenoid valve wired to my furnace. I plumbed well water through the evaporator coil and directed the waste water outside to an outdoor faucet which, in the summer, is hooked up to soaker hoses to water the flower beds... The plants like the warmish water and, while not terribly efficient cooling, it does manage to keep the inside of the house below 22C when outside temps are over 30C... The house has a lot of solar heat load due to big windows with mountain views on the west side and even with awnings up, would get excruciatingly hot without some cooling assistance... My only operating cost is the electricity to pull the water out of the ground
I could probably make better use of the waste water by sprinkling it on the roof before collecting from the eaves and doing drip irrigation on the flower beds, but that will be a project for another year.
(This is in Southern Alberta)
Spongebog Squarepants called... (Score:1)
...and wants his water back.
Patrick........!!
Fortunately no unforseen consequences (Score:2)
I mean, all this heat is just going to vanish into the environment, right?
Re: (Score:2)
So your answer is... turn it off? Turn it all off? Sit shivering in the dark?
No, wait, that's not enough. Sitting shivering in the dark is STILL dumping heat into the environment.
DIE! Yeah, that's the ticket. No metabolism means no metabolic waste heat. (Assuming you'll forgive us our methane and CO2 output as we decompost. I promise it's only momentary. Humanit will no longer be responsible for ANY environmental warming within a few weeks after we've extincted our entire race to make you feel better.)
You m
Re: (Score:2)
Basic answer?
It's too late- we are screwed.
Real answer? Get the human population down to about 3 billion and live with a lot of freedom and prosperity. Continue down the same path and look at rationed water, cramped living quarters, lower quality food. And when we do have something bad, we'll have a tremendous die off.
Look- wind power has now been shown to reduce the winds (and affect the climate). And seriously... DUH? You are extracting huge amounts of energy from the wind- it's going to affect the w