Norway Energy Issues and Ammo Production

[Darwin]

You should see the server farms that run these things. The entire roof is cooling fans and heat exchangers. The numbers here actually make perfect sense.

What’s a real red flag to me is a Chinese company that is at the beck and call of the Chinese government, has disrupted western ammo production close to one of Chinas most offensive allies - Russia.

David

I had no idea of the scale of the server farm contraptions. Thanks David.

 

[rijndael]

I had no idea of the scale of the server farm contraptions.

Our facility is setup for ~12 MW of power from the various utilities. 40% of is used up in HVAC.

 

[Darwin]

No I understand the diesel gen sets powering electric car chargers. I guess Electric Armageddon comes after the ukabook war on the worldwide clown show.

 

[JMayo]

I NEVER wanted to see tictok until now. They have dancing hotties??

Yep. I don't watch porn at all anywhere.....
but the scantly clad honeys definitely brighten up your day..... or make you cry in your beer!

 

[3Sigma]

One blade ... a blade server? What damage would it cause? The max heat output from it 'overheating' is determined by the power supply's maximum draw, assuming it's all converted to heat.

We have dozens of blades in a rack. Our racks have an average maximum power of 2 x 8kW / rack. One blade (in a rack) overheating is not noticeable and no one but the server owner would care.

Apologies if I am getting it wrong. My perspective is from the design of the suppression systems, so a worst/best case view of an exothermal event. Not an electronic power draw, but actual fire. Obviously, the sooner you can detect, the easier to isolate and protect, the lower the impact, and the lower cost. I'm sure there are systems designed to detect and cut power before anything gets to the fire system's sensors. It's only a part of the safety systems.

But, the risk of fire is there due to the heat generated by these systems. Heat that has to be removed by very big cooling systems. Systems that need a lot of power to operate. Hence the impact on the power grid mentioned in the original post.

 

[Brians356]

Easily? Every data center I've worked in, dozens of them, had significant fire suppression systems. They're often Dry Pipe, FM200 or 1230. No one wants hundreds of millions of dollars of equipment burning up. I've even been in one when equipment arc flashed.

If a suppression system is activated, then a fire has triggered it. The fact that any fire is quickly suppressed does not mean fires cannot start easily. Semantics, sure, but you both may be right. The question really is: How often are suppression systems activated?
-

 

[rijndael]

. The question really is: How often are suppression systems avtivated?

IME, they're not often at all.

FWIW: I've been mostly in data centers since 2003. I work in one daily.

All of the fire alarm activations I'm aware of were from pipe welding and copper solder joints when doing building maintenance.

The arc flash event was the only event that came close to needing fire suppression, and that was in the power distribution room, not the server area. I've seen one power supply on a firewall start smoking, but someone walking by saw it and pulled the power cord.

But, the risk of fire is there due to the heat generated by these systems.

IMO, this is a non-issue. The risk of fire that deserves attention is in the building's power infrastructure. The step down transformers, UPSs, transfer switches, etc.

 

[Brians356]

IME, they're not often at all.

FWIW: I've been mostly in data centers since 2003. I work in one daily.

I knew that answer, since I'd worked in data centers myself, present at the invention of the first big modern ones. My territory in the industry 20+ years ago included Silly-con Valley before the Dot Com crash. Microsoft was one client. I helped design one of the very first parallel computing server farms (NOAA Forecast Simulation Lab in Boulder, CO) long before the term "blade" meant something other than a knife or a plucky skirt-chaser. Back then there were racks of 1U "pizza boxes" and 256 of them were awe-inspiring.
-

 

[rijndael]

Back then there were racks of 1U "pizza boxes" and 256 of them were awe-inspiring.

What's old is new. We're not going blade frame chassis units anymore, it's back to 1U boxes.

 

[jackieschmidt]

Everyone knows that TicToc is more important than national defense!

AOC and her Democrat Compadres sure think so.

 

[Brians356]

What's old is new. We're not going blade frame chassis units anymore, it's back to 1U boxes.

Well, after all, I've been retired for some years now. I leave such vexations to you young blades.
-

 

[Dusty Stevens]

Just for a little comparison, the feeder on the left is running a new crypto farm. It is at 25% capacity today and its one of the new liquid cooled ones so it runs way cooler than some of the others. It will be running 1800 amps this summer. The other 2 feeders feed a whole town and the outskirts, probably 10,000 customers or so. We give them such a good deal they pay to upgrade our substations.

 

[FNSafari]

AOC and her Democrat Compadres sure think so.

its a drug to keep the masses focused on general silliness rather than the issues they should be concerned with.

 

[sw282]

Como and data centers have always been power robbers. l remember reading about the worlds' first super computer England during WWll. ENIAC it was called l think. Built to crack the German Enigma Code. ENIAC used almost as much electricity as half of London. All vacuum tubes then, No transistors or chips

 

[xr650rRider]

Bitcoin mining facility pretty sure they are bankrupt now.

 

[dhagerty40]

after reading this thread, It has made feel Very Old! The data centers I worked in as an electronics technician were during the 70's and 80's and last one I worked in was 1992. Back then, Microelectronics and Solid state memory was in it's infancy. I just remember that we had to keep the Mainframe room at a constant 68f. And now, a present day laptop has 10 times more processing power than that Mainframe computer had!

I might be one of the Few techs that can still troubleshoot vacuum tube systems down to component level!