Slashdot Mirror
Liquid Blade Brings Immersion Cooling To Blade Servers
1sockchuck writes "In the past year we've seen several new cooling systems that submerge rack-mount servers. Now liquid immersion cooling is coming to blade servers. Liquid-cooled PC specialist Hardcore Computer has entered the data center market with Liquid Blade, which features two Intel 5600 Xeon processors with an S5500HV server board in a chassis filled with dielectric fluid. Hardcore, which is marketing the product for render farms, says it eliminates the need for rack-level fans and room-level air conditioning. In recent months Iceotope and Green Revolution Cooling have each introduced liquid cooling for rack-mount servers."
Although it's good we don't have them. I'd probably get fired when they find a rack of production servers running at 4.6GHz.
finally some good news for Joe the Plumber.
Do we really NEED liquid cooled servers in datacenters? Is this just our feeble attempt to validate Moore's Law despite diminishing returns on smaller process size and core multiplication...?
What the hell am I talking about? Of COURSE we need them!
If the only way you can accept an assertion is by faith, then you are conceding that it can't be taken on its own merits
How hard is it to say; change a disk in one of the submerged nodes ? or fix a loose ethernet cable ? If the nodes are separated in compartments, and you could isolate and drain one while servicing it, this would be really nice indeed.
Doolittle :
Bomb no.20 : To explode of course.
Mafia rap + enterprise computing = Samir?
Last line of a job add:
Having a diving certificate is definitely an asset.
I'd like to be under the sea in an octopuses data garden in the shade...
Immersion liquid cooling is something I have done in the past, and that is all well and good, it is after all HOBBY level tech.
For commercial level tech it isn't even a joke, imagine opening the bonnet / hood of your new 2010 car and finding a big tub full of water with the engine immersed in it.
Internal combustion engines have had closed circuit internal liquid cooling circuits for decades, and frankly computers and electronics have had closed circuit internal liquid cooling circuits for decades too.
Think backplane technology and hollow main boards, the liquid coolant flows through the hollow PCB, and mates and either side with the "backplane".
All the advantages of liquid cooling, and almost none of the disadvantages of liquid cooling.
Air cooling has one great advantage, "leaks" don't matter. Provided you have sufficient mass flow you can leak air all over the place.
Older internal combustion engines didn't even have forced circulation in the closed loop liquid coolant systems, they used thermal syphon, much like the space between the racks.
The salient fact here is you have to design in the cooling circuit at the engine block / PCB mechanical design stage, until and unless you do that you are going to be dealing with some god-awful heath-robinson kludge, like fitting an old "stationary engine" (google it) into a 2010 Dodge rolling chassis.
Instead of a 50 buck case containing a 100 buck mobo, you end up with a 100 buck case and a 200 buck mobo for closed circuit air cooling, or a 200 buck case and a 400 buck mobo for closed circuit liquid cooling, and these prices are for large volume manufacture with full economies of scale.
Now go back to your Dodge dealership and take in two 2010 rolling chassis for the annual service, one is running a bog standard cummins, the other is a kludged up stationary engine, and ask the mechanics which one will be more expensive to service.
Closed circuit liquid cooled electronics are not new, it is routinely used in avionics, which of course means that you can back 200 watts of thermal rejection (a modern desktop computer) into a package the size of an iphone, and run it flat out 24/7.
But it costs.
Unless you are in Hong Kong then the cost of land per acre is cheaper, and air is free, and leaks don't matter, and the coolant doesn't cause shorts.
The only other advantage of liquid coolant is it is much quieter, but even so, you can cure that problem by making everything bigger to accommodate much larger passive heatsinks.
http://hackedgadgets.com/wp-content/_hs2.JPG for example, this stuff is extruded and bought by the metre, it doesn't have a failure mode.
http://slashdot.org/~GuyFawkes/journal
Switch to ARM, guys
http://www.greenm3.com/2009/09/arm-pushes-performance-per-watt-announces-2ghz-multicore-designs.html
Go where it's wetter, you know that it's better!
Science advances one funeral at a time- Max Planck
I don't know why you'd have to immerse the entire blades in cooling.
Couldn't you just use tube liquid cooling, found in many enthusiast machines? You could make one pump/heat exchanger per blade enclosure, with a custom valve/fitting connecting the blade to the enclosure to pass the cooling around. I'm sure a clever engineer could easily come up with a design for redundant heat exchangers/pumps in the enclosure, even hot swappable.
The benefits of liquid cooling (low noise, lower temperatures) with less or no mess. I'd also imagine it would be cheaper than this method as well.
This sounds awesome and I want one but I'm not gonna put it in my budget until Sun(Oracle), IBM, HP or Dell are selling it
Is that similar to a Light Saber?
> marketing the product for render farms
I used to have root on a renderfarm with a few thousand cores and this is exactly the type of fiddly tech you want to avoid when you scale-out. Many renderfarms are located at multiple sites, mainly due to legacy but also for convenience and resilience. This means the majority of a renderfarm may be co-located where land and electricity is cheap and rely on high bandwidth connections to offices where transport and talent is plentiful. I know of two renderfarms which are split in this manner.
It means that really big renderfarms are co-located with shared cooling facilities. This makes fancy cooling methods very risky and unnecessarily expensive.
Does anyone make a liquid cpu heatsink, something to slap over your cpu (and seal)? Seems like that would be a nice innovation if it could be kept sealed.
This is a great step in the correct direction. First we had swords with bronze blades, then iron blades. Now we have liquid blades.
The next step should be the plasma blade and the ultimate goal will be the lightsaber.
Don't fight for your country, if your country does not fight for you.
I wonder what sort of affect this 'liquid dielectric' has on the hardware itself. Lots of problems associated with high frequency networks are capacitive couplings. That is to say that there are virtual capacitors between copper traces on the circuit boards themselves simply because they are so close together. Any first year EE should know that as the frequency of a signal increases, any capacitive elements inside the circuit behave more and more like short circuits (Tada! Filters!). It seems to me that were removing air as the parasitic capacitor's dielectric and replacing it with something else, probably something that has a permittivity much higher than air's (~1).
Cray Supercomputers were doing this a quarter of a century ago. Fluorinert to the rescue!
"To those who are overly cautious, everything is impossible. "
Immersion cooling makes sense from a lot of perspectives. However, there is one enormous problem. That is that chips aren't rated to work when immersed. You will have to get the companies who make the chips to specify that the packages the chips are in will are designed and safe to use in liquid. And they're going to charge you a lot of money for that. Frankly, it'll make the device non-cost effective.
You can just omit this step, and throw something together with air-rated chips and hope it works anyway. But you don't really want to depend on such a system for mission critical services.
http://lkml.org/lkml/2005/8/20/95
For special and expensive systems such ideas may be useful, but it is hard to beat air for commodity systems because it is free, there is plenty of it, it cannot spill, it does not put excess load on the floor, it is inert (to electronics), and it can be easily recycled.
HPC for Primates. Read Cluster Monkey
Than re-fitting all your racks with some carbon fiber advanced material. Surely cooling with a pumped liquid generates less vibration in the air than having a bunch of high-speed fans in your case.
Maybe this will result in better hard drive performance :)
Water is dielectric by itself.. Electric conduction is because of all the mixed salts in it. Perhaps they are just selling distilled water? :)
Anyway, you would have to change the liquid from time to time, because some dust or anything might get into it and dissolve, thus making it conductive..
... killing machine comes handy both at home and at the office. Call now, and order directly from the creator Cyberdine corporation.
That “dielectric fluid” is really nasty stuff. To give you an idea, how nasty it is, I have an example for you:
Say you have attached your mouse to the computer, and sealed everything off with hot glue and everything. Now you fill the thing with that oil.
Then that stuff creeps trough the connector, trough the inside of the cable (between the wire and the plastic), up to your mouse, out your mouse, over all your table and everything, down to your floor and trough the whole room. Unless you stop it first.
Now add dust, and your desk remembers you of something that’s tarred and feathered.
Nasty nasty stuff. And hey, the heat of course doesn’t magically disappear. So you still need to cool the oil or the surrounding air somehow, and keep the oil moving. Or else you are much more prone to heat accumulation.
So all in all, it simply isn’t worth the effort. By the way: What’s so bad about fans anyway?
Any sufficiently advanced intelligence is indistinguishable from stupidity.
Just friendly crustaceans
GENERATION O98346: The first time you see this, copy it into your sig and remove a random number from the generation. T