[Beowulf] bring back 2012?

Prentice Bisbal pbisbal at pppl.gov
Mon Aug 22 08:14:01 PDT 2016

On 08/20/2016 10:03 AM, Stu Midgley wrote:

>     I don't need to try to do that, because that's the job of the
>     engineers at Asetek, CoolIT and their OEM partners. I've looked at
>     their solutions every year they've been as SC, and the packaging
>     is quite neat, and included heat exchangers for the RAM to keep it
>     cool, too.  I sure as hell wouldn't want to engineer and build the
>     system myself, but I'm sure glad there are others doing it, and I
>     have a lot of confidence in their work.
> Spoken like someone who doesn't have to take responsibility or pay for 
> their solution.  I work in the private sector and have to satisfy 
> customers.  If a solution doesn't work, or has continual problems or 
> is delivered late, I have to answer for that and our company suffers.

Spoken like someone who would rather attack their opponent in a debate 
than stick to the issues of the debate. As I stated earlier, I have a 
degree in an engineering discpline that I think makes me more than 
qualified to evaluate the design of these different systems, and choose 
the best one. I don't know about where you work, but I've never been in 
a job where employees aren't held responsible for their decisions at 
work and aren't at risk for losing their job if they make a bad 
decision. In fact, New Jersey, where I work, is known as an 'at-will' 
state. This mean I can quit without reason, and my employer can fire me 
without reason. If that isn't enough reason to make good business 
decisions, I don't know what is.

>     Now my criticisms, which apply to mineral oil immersive cooling only:
>     * The stuff is messy. I can't imagine removing components from a
>     server without them slipping out of my hands. I can't even imagine
>     getting one from the tank to a workbench without getting oil
>     everywhere. I'm sure you can clean it off with a lot of electronic
>     parts cleaner, but how good for the environment is that?  I see
>     this stuff gaining more market share when we get to 'disposable'
>     servers with ARM-based SoCs, so cleanup isn't even an issue.
> I work with the stuff as do our local SGI engineers and it isn't that 
> big a deal.
>     * Liquids are dense! Immersive cooling in mineral oil adds a lot
>     to the load of the data centers raised floor. Yes the tanks are
>     about the size of a rack flipped on it's side, but even the
>     densest servers are still mostly air by volume. I imagine many
>     data centers would need to upgrade their floors before going to
>     this type of cooling.
> Wrong.  Prior to immersive cooling, we were running SGI chilled water 
> racks.  A 40RU tank with nodes comes in at about 1600kg's... an SGI 
> chilled water rack with nodes comes in at about 1400kg's.  The SGI 
> rack has 1400kg's in less than 1msq whereas a tank is 1600kg's in just 
> over msq... it is actually better for your floor.

I will concede this point. What you say sounds plausible, and I don't 
have any numbers to counter with.
>     * Since the "racks" are on their sides an only go up about 3 - 3.5
>     feet, it's not a very efficient use of space. We often talk about
>     data center space in square footage, but it's *volume* that really
>     matters, and I don't think mineral oil cooling is very space
>     efficient.
> Wrong.  If you take a rack and the clearances of the back and front 
> required to service nodes (400mm at the back 900mm at the front), you 
> take about 2sqm per rack... which is less than our tanks currently 
> take.  We can get MORE density with an immersive solution than with racks.

I'm still skeptical. You still need space for people to walk through the 
tanks on one side, and cabling on the other side. I can see how the 
cabling aisle takes up less space than a hot aisle, but I'm sure it must 
still be wide enough for someone to walk through so they can 
install/remove/repair cables, unless you do something completely 
different in this case and run all the cables in cable trays directly 
over the tanks. Is that what you do?
>     * What if one of those tanks leaks or a pipe bursts? I'd rather
>     clean up water or Novec. Yes, there are electrical dangers when
>     that happens with water, but that's what circuit breakers and such
>     (GFCI, AFCI, etc.) are for.
> Water is actually worse than these fluids.  If you get a leak, you use 
> a wet-vacuum and suck it up and put it back into a tank.  You don't 
> have to worry about any electrical components or other mechanical 
> stuff.  The fluids are easy to clean up, wipe down and get on with life.
>     * What happens when you decommission hardware that's covered in
>     mineral oil? How do you clean it up? Will a recycling center take
>     electronics covered in mineral oil.
> When you decommission nodes, you throw them out like you do with all 
> old hardware.  The fluids are non-toxic, biodegradable, have a clean 
> MSDS and nodes/components can be disposed of clean or oily.
> -- 
> Dr Stuart Midgley
> sdm900 at sdm900.com <mailto:sdm900 at sdm900.com>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.beowulf.org/pipermail/beowulf/attachments/20160822/1412d699/attachment.html>

More information about the Beowulf mailing list