On 08/22/2016 11:22 AM, Stu Midgley wrote:
While the risk of an explosion is a certainly a theoretical
possibility, In practice, the risk of this is virtually
non-existent for a variety of reasons.
With water, the processors and other heat-generating components
would fail from the heat before the boiling point of water is
reached, so there would be little to no generation of water vapor
that could lead to an explosion. Also, any heating/cooling system
with water would be designed to included an expansion tank to
account for the thermal expansion and contraction of water. There
are millions, if not billions, of homes and businesses in
existence with hot water heating systems, yet, I've never heard of
any of them ever exploding.
With Novec and other two-phase systems, the gas phase is
compressible, meaning it can store energy like a spring,
preventing or minimizing the case risk of an overpressure
situation rupturing the vessel. All that is required for this to
be used safely is an adequate volume for the gas, so that is has
excess 'capacity' to be compressed. This simple design is what
allows 20-pound propane tanks to be used all over America (and
probably other countries) to fuel gas grills and be left out in
direct sunlight all summer long, and be stored directly under the
heat-producing burners. If those tanks were filled to the top,
they would explode in those conditions, but but leaving about 1/3
of the tank empty, the risk has been virtually eliminated. This
was actually a top we spent a lot of time discussing in my
Chemical Engineering Safety class in college.
This also applies to the tanks storing liquid nitrogen, liquid
oxygen, and other gases/liquid stored well below their boiling
point. Tanks of these substances can be found throughout the world
in industrial and laboratory environments, yet explosions caused
by them are quite rare. When they explode, it's usually because
someone who didn't know what they were doing overfilled the tank,
or the ambient temperature exceeded the designed safety margins
through some other catastrophic event. (structure file, etc).
Finally, all systems where this is a risk would have plenty of
safety features to prevent this. My gas water heater at home has a
simple temperature/pressure switch to safely discharge excess
pressure/temperature event. These are cheap, readily available
items that you can buy at any local hardware store. I also have a
steam heat system in my house. In the early days of steam heat, it
was not unheard for a steam boiler to explode with devastating
results, but just to some simple design elements (Hartford Loop)
and basic mechanisms (low water cut-off valve, pressure relief
valves) have virtually eliminated this risk.
Before I got in to HPC as a profession, I was a process control
systems engineer. My companies specialty was control systems for
boilers for power generation. The pressures of these systems were
much higher than what we're talking about here. Our systems had
plenty of pressure sensors, release valves and failsafes.
Incorporating any of these safety elements into a cooling system
like this is trivial, and I'm sure the vendors who sell such
solutions have already done that where appropriate
But if you have 40kW of gear still running, your not storing the
liquid in the sealed container well below boiling point - its actually
the opposite you are running at or just above the boiling point. Even
if you take the approach "our systems will shot down if we loose the
external cooling circuit)... that still takes time to recognise and
shutdown... mean while your systems are pumping heat into the tank.
Any failsafe mechanism would be designed to shut the system down with a
wide safety margin.
Again, with the boiler example, this isn't the sort of behaviour you
want in a computer room. You don't want this stuff venting... and
also, try and get a permit to operate such a system in an existing or
new facility.
In reality, I don't think any of this stuff would need to vent, or even
have vents, I was just explaining that there are a multitude of ways to
mitigate the alleged explosion risks based on my Chem Eng. education and
several years of experience as a practicing Chem E. I'm afraid I may
have gotten a little carried away and distracted from the 'in actual
practice' argument I was trying to make.
If you look at the IceTope literature, their Novec loops are designated
'low pressure' so, I'm assume there is plenty of excess capacity in them
for pressure to build up safely.
With a non-phase change solution, this isn't an issue.
Novec is a common fluid used in fire suppression systems in
computer rooms... so you shouldn't have too much of an issue with
the flammability of the fluid, but it is an entirely different
issue to get them to warrant the solution.
If it's already used in fire suppressions systems, where it's
going to deliberately be sprayed into the atmosphere, I don't see
how a system where it's designed to be completely contained would
more of a safety issue, but this could be a local practice issue
(USA vs. Australia laws, etc.) In the event of an emergency
venting, as stated above, the vents are hooked up to exhaust
piping so that that gases released are piped away to a location to
where it can be safely released to the environment, so a venting
event would pose no risk to the occupants of the data center. This
is done all the time at chemical plants. In fact, venting like
this happens quite regularly in those environments, but I honestly
don't even see this type of venting being needed in a system like
this.
If you look at the US laws around this (I've only been through the
Houston documentation - but I assume all states are roughly the same)
their is a massive different between holding a few hundred litres of
fluids (which is what is in a fire supression system) and say 30000L,
which is what you'll have in 30 tanks.
I don't think you'd anywhere near that volume of Novec in a system. The
cooling systems volume are only partially filled with Novec, since you
want to have a low pressure and room for the gas to expand. This system
only works if the Novec can easily change state. Since the change of
state absorbs a lot of heat, you need a lot less of the fluid to have
the same thermal capacity.
Going back to my home's steam boiler example, My boiler only holds a
couple of gallons of water. If I had hot-water radiators or hot-water
baseboards, I'd have you use a lot more water.
Also, the Novec is only in the primary loop. It condenses on another
heat exchanger which transfers the heat to a secondary loop which
carries the heat out of the data center, so not as much Novec would be
need as you might be assuming.
From what I recall, the Green Revolution system is a single loop filled
with mineral oil all the way to the external heat exchangers.
Perhaps the final nail in the coffin of the Novec solutions... it
has been on many different booths at SC for many years... now go
and try to find a vendor that will actually sell you a solution...
I guess you never stopped by the Icetope booth at SC, then.
They've had solutions on the market for several years now, and
have had booths at SC for several years now, too.
http://www.iceotope.com/
I've seen them, I hadn't realised they were using Novec. Even a 5min
look through their website doesn't make that clear.
Yeah, I noticed that when I checked the link, and I was very
disappointed in that, since if I couldn't prove it, it wouldn't help my
argument. They used to be much more 'vocal' about that. Icetope worked
with University of Leeds to develop and test their system. If you
remember, there was member of this list involved in that work who would
occasionally post updates on their work here.
Which brings us to another topic... and that's price. The icetope
stuff looks very very custom... and thus very very expensive. I can
purchase gigabyte or supermicro equipment already for the fluid we are
using (they modify the power supplies, leave the thermal paste off
components etc)... no modifications at our site necessary - and
relatively cheap (is any HPC gear cheap?)
Damn it! I can't argue this point at all. You are absolutely correct
that Icetope's design requires purpose-built cases. I think with enough
volume (no pun intended), the cases could be manufactured almost as
cheap as existing air-cooled cases, but until they get that kind of
volume, I'm sure they're more expensive another con is that I don't
think any of the big OEMs (Dell, HP, etc) are working with them at the
moment, limiting the options.
--
Dr Stuart Midgley
sdm...@sdm900.com <mailto:sdm...@sdm900.com>
_______________________________________________
Beowulf mailing list, Beowulf@beowulf.org sponsored by Penguin Computing
To change your subscription (digest mode or unsubscribe) visit
http://www.beowulf.org/mailman/listinfo/beowulf
