Data center water cooling and other thoughts.

Today I’m going to talk about some of the green data center techniques that are getting a lot of airplay but appear to have serious deficiencies with them. Or is it just me?

The first point comes from reading about using Sun rear door cooling and previous articles on IBM rear door water cooling. It might just be me, but it would make a whole lot more sense to put your cooling in the intake air path where it will do the servers some good, not the outlet path after the air has gone through the servers. Intake air should always be as cool as possible.

Iceotope has an interesting take on fluid cooling by immersing the entire motherboard in coolant. On the other hand, this approach requires 3 separate cooling loops (server, rack, main) with losses at each point. Couple that with increased complexity and the potential for failure at any point (lose a rack level pump and whack the whole rack?) and I have a hard time seeing this being practical.

Last idea on the block tonight is the idea of burying micro nuclear reactors in old missile silos. Sounds like a great idea until you consider that most if not all obsolete missile silos are in areas without the water supplies required operate a nuclear reactor.

Maybe it isn’t just me after all.

Vern, SwiftWater Telecom


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s