A new wave in data center cooling?
Chilling with Phil Tuma
Webcast During the SC09 HPC show last November, I walked around with a small video cam and did some interviews from the floor, posting them to The Reg in lieu of blog entries. It was a lot easier than actually writing blog entries and quite a bit of fun too.
One of the booths I visited was displaying their solution to removing system generated heat – basically a rack of blades immersed in mineral oil. It was a different take on liquid cooling – like liquid cooling with a vengeance. The blog attracted a bit of attention and garnered some interesting (and funny) comments.
A few weeks later, I received a email from Phil Tuma of 3M. He told me that they were working on something along those lines and asked if I was interested in learning more. I was, of course (talking about new technology with others is a lot more fun than sitting alone typing about it).
During the briefing, they laid out their approach, their technology, and their proof points. What impressed me was the experience that 3M has in this area and the obvious rigor they have applied to their solution.
Of course, the proof is in the pudding and their pudding isn’t quite cooked yet. But the test they shared with me showed them holding constant temperatures while cooling a 8”x7” board populated with 20 200 watt ceramic heaters (4kw total). They did this using a liter of their special fluid and a gap of .16” (4mm) and .28” (7mm) between the edges of the enclosure and the heaters – very dense packaging indeed.
In this webcast, Phil Tuma and I talk about the drawbacks of air and traditional liquid cooling and take a look at their 2 phase immersion approach. To me, there is a lot of promise in this technology. The major hurdle right now is that it will require a fair amount of engineering to transform system boards to take advantage of the massive cooling effect from immersion.
On the customer side, implementing immersion will necessitate some extra datacenter plumbing to run water lines from tank to tank and to take waste heat off for recovery or discharge outdoors.
Take a listen to the webcast and let me know what you think. I think this may raise a lot of questions and I’d like to have Phil back to answer them and provide more information. If you have a question or comment, drop it in the comments box under this article and I’ll use ‘em in the next webcast. ®