IBM Pushing Water-Cooled Servers, Meeting Resistance

judgecorp writes "IBM has said that water-cooled servers could become the norm in ten years. The company has lately been promoting wider user of the forty-year-old mainframe technology (e.g., here's a piece from April 2008), which allows faster clock speeds and higher processing power. But IBM now says water cooling is greener and more efficient, because it delivers waste heat in a form that's easier to re-use. They estimate that water can be up to 4,000 times more effective in cooling computer systems than air. However, most new data center designs tend to take the opposite approach, running warmer, and using free-air cooling to expend less energy in the first place. For instance, Dutch engineer Imtech sees no need for water cooling in its new multi-story approach which reduces piping and saves waste."

Any prediction over ten years is null and void

[Meshach]

These kind of predictions always remind me of Bill Gates asserting that "640 K should be enough for anybody."

Hardware and software faces change so fast; who has any idea what will be required or available in even ten years?

Re:hmm..

[Anonymous Coward]

This is IBM... Do you _really_ think they'd design it in such a way that you'd have to take down the whole thing to fix a small section?

You wouldn't have one long pipe running to all of them, with no way to shut off segments/individual nodes.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Re:4000 times?

[Chrisq]

Just try this. See how long you can stand naked (OK wear some running shorts) in air at 5 degrees centigrade. Probably fifteen minutes standing still or indefinitely if running.

Now see how long you can stay in water at 5 degrees centigrade. For most people it would be less than a minute - you may not even be able to get in.

Re:Ratio of specific heat capacities

[SlashWombat]

Yes, but eventually, all that heat ends up in the air anyway ... the water is only the middleman. Water is actually probably the most efficient coolant around, however, the latent heat of evaporation means it works best when it is boiled off the surface to be cooled. This is not exactly ideal for a semiconductor, although it might be okay if the water was in direct contact with the silicon. (Silicon junction temperatures must be kept below 360 degrees Celsius.)

Re:Not for all

[stephanruby]

I think for now, many companies are perfectly ok with air cooling solutions. Besides, it's much safer to have air-conditioning and fans than some liquid flowing.

If some companies can make fridges that do not leak coolant. I'm pretty sure IBM can make mainframes that do not leak their coolant either.

Re:Any prediction over ten years is null and void

[Nefarious Wheel]

It's much easier to predict the past, however, if you've been paying attention. Early computers blended their cooling system with the heating system of the surrounding building. They were sometimes designed together that way.

Re:hmm..

[Nefarious Wheel]

Sort of depends on where the water's coming from, doesn't it? I remember once when the water piping in an old -- think it was a 360/95 or some such oddity -- failed (yes, it was a looong time ago) and the area under the false floor flooded. This was before Ethernet and the floor was a rats nest of individual terminal cables (not from the 360) -- hundreds of them, along with power cabling. The real problem surfaced (so to speak) some time later, when the actual rodents who did make a rats nest of it displayed the properties of dissolved urea and the effect of said resultant acid on the pre-teflon (I did say it was a while ago) cable insulation.

Needless to say, it was a rat shit situation, and I was never more glad that I'd gone down the software track. Nobody wanted to get anywhere near the network guys for a while.

Water cooling should not be a sop to consumption

[niks42]

Whatever the technologies that are available, we shouldn't as a society use them to be wasteful. If we could still efficiently cool a power-hungry, inefficient processor complex using a liquid rather than air, it doesn't make it a good idea. It would be better to design a data center that didn't require such amounts of energy in the first place.

This reminds me of recycling schemes that make people think it is OK to overpackage goods in the first place.