Data Centers And DC Power

mstansberry writes "In the final article in a series on the price of power in the data center, IT pros weigh the pros and cons of direct current-powered servers. A limited number of companies make servers with the power supplies removed with DC power distributed to multiple machines from a single unit. It saves power by skipping an extra conversion from alternating current (AC). Telcos have been using this method for years, but some data center pros are leery of taking on the new systems. It's not something people are familiar with and if they break down, you have to hire a specialized engineer to come fix them. But if they're saving even half of what they're reported to save on the electric bill, companies could afford to hire the engineers." We've reported on previous articles in the series.

Comments

What about houses?

[jolyonr]

I've always wondered (from a non-technical point of view) whether there was a benefit in having our homes wired up with two sockets (or maybe a 5 pin mains plug) giving standard AC voltage and a low-current DC voltage as well (12V?). So many devices only need low voltage, wouldn't we all benefit in having a power system in our houses in this way?

Jolyon

Re:What about houses?

[Crispy Critters]

"Since when was non-ionizing EM radiation dangerous?"

It isn't.

There are three parts to assessing something like this.

First, is there some physical method consistent with known science by which injury could occur? In this case, no.

Second, do controlled experiments on human or lab animals show an effect? In this case, no.

Third, do statistical studies show a correlation? In this case, not when they are done competently. The problem is that a lot of studies have been done by people who don't understand statistics. They think that everything should be exactly average. They look at a bunch of populations, find 20% are more than one standard deviation above average, and try to find some explanation for this.

Sort of like Dilbert's PHB getting angry when he finds out that 40% of employee sick days are taken either on a Monday or a Friday.

Re:What about houses?

[Crispy Critters]

I think it arrogant to assume artificial electric fields cannot have an effect on their proper operation.

Of course it would be. That is why I explained things the way I did. There is a difference between saying "It has been proven that 60 Hz fields do not affect humans" and saying "There is no evidence that 60 Hz fields affect humans."

We can speculate all day about what could conceivably affect a human body, given the limits to our understanding. But that leads to tinfoil-hat country unless we look for some evidence in support. The fact is that no evidence shows that humans are hurt by 60 Hz EM fields.

Maybe some will appear tomorrow. Who knows? Certainly the study of how the body uses electrical signals is interesting, and maybe one day we will learn that a lot of the current accepted wisdom is wrong.

The question is whether the mere speculation that 60 Hz fields can harm people, with no supporting evidence, should cause us to condemn every house within a mile of a high voltage power line, install Faraday cages around every bed, and sue the power company over every case of cancer.

how does it save a conversion

[Anonymous Cowpat]

ac goes into data centres, systems run on dc. Either it gets distributed to each computer as ac and converted in a medium-sized box in the back of each system, or it gets converted in one big box and distributed to the systems as dc.
The question is of the efficiency saving of doing all the converting in a big box against the efficiency loss of piping it around the data centre as dc, and wether you get a large total net saving (which I suspect that you do, since even inside the data centre, it's not going far)

Re:how does it save a conversion

[qwertphobia]

Power gets converted to DC anyhow to keep the UPS batteries charged. If the lights go out, the DC from the batteries is converted back to AC to go to the power supplies and back to DC inside each system.

No, it doesn't take as much power to keep the batteries charged as it would to run the center off DC, but that's not the point. Anyone with a large UPS already has a beefy AC/DC and DC/AC conversion system in place.

I would also assume one large converter / power supply would be more efficient for power and heat than hundreds (in my data center) or thousands (in a big one) of little power supplies. Any thoughts on that?

DC power for data centers

[dawggy_daddy]

trouble shooting and correcting DC power is simpler than working with linear power supplies. Unfamiliarity is the problem, not the technology.

correction

[Dun Malg]

It saves power by skipping an extra conversion from alternating current (AC).

No, it saves power by having one large, very efficient power supply do the conversion to DC rather than several dozen small ones. Either way, there is still only one conversion: 110AC -> 12/-12/5vDC.

specialized engineers?

[M. Baranczak]

I don't see what the problem is. Each box would be built exactly the same as before, except it wouldn't have a power supply unit - just a straight wire from the power socket. And the central power supply is just a big AC/DC converter - most EEs should already be familiar with those. There's not really any new technology here, it's just slightly re-arranged.

What everyone seems to forget

[jeffmeden]

Is that DC power is naturally unstable. As loads fluctuate, the conversion and distribution system can change dramatically and result in very unclean power. If you are proposing to ditch AC in the server room and run DC from the UPS hardware directly to the rack, you will need to add in a lot of hardware to guarantee that the servers get exactly the voltage they need. This hardware will probably be less costly and wasteful than the AC systems currently in use, but they will also be more proprietary and (in the short term) more expensive to buy into. This is not the magical solution many envision, but it has a good future since transistor technology is getting a lot better and hence voltage management will be easier and easier as time goes on. The opportunity to move the conversion heat away from the inside of the server allows for better heat management, since you can let a transformer/transistor power system toil away only cooling it from the air duct on the roof and save the crisp AC for the servers.

How does it save power?

[harlows_monkeys]

OK, electricity has always confused me, so I'm probably being stupid here, but I don't see how this saves power.

Assume that an AC-to-DC conversion causes a loss of 10% (just to have a number).

If we bring in AC, convert it to DC in one location, and then distribute it as DC to all the computers, we've lost 10%.

If we bring in AC, distribute it to all the computers, and convert to DC at each computer, we lose 10%. The conversions are independent and parallel, and so the loss is not additive. (After all, if we have 10 computers, it doesn't mean we are losing 100% of the power). I can see how we might save money, as we no longer would need a complicated power supply at each computer. Also, we wouldn't have a hot power supply in each computer, and this could reduce cooling costs. But I don't see where the power savings comes from.

Saving a conversion step isn't the issue.

[RockyMountain]

The slashdot story intro implies that the advantage of DC is that you
save a conversion step. Well, maybe you do, maybe you don't, but
counting the number of AC-to-DC and DC-to-AC conversions is very
misleading.

Converting 50 or 60 Hertz to DC is much more costly and less efficient
than converting in either direction at a higher frequency. Low
frequency rectification requires large filter capacitors, complex and
expensive inrush current limiting, and active power-factor correction.
  By doing that front-end work in one place only, preferably from a
3-phase source, you save power and increase reliability. You probably
still want multiple 50/60Hz to DC rectifier stages, of course, but now
they can be in parallel (for redundancy), rather than each one
downstream of the other where a failure of either one will bring down
the system.

Just because you're distributing DC to the racks, doesn't mean you
don't have to convert it again. It typically gets converted to AC and
back to DC at least once, usually twice before it reaches CPU and
memory chips. That's equally true in data centers that distribute AC
or DC. The fact is, memory and CPU devices want very low DC voltages
and very high currents. To make matters worse, not all parts of the
system want exactly the same DC voltage, you almost always have to
have multiple supply rails. You can't distribute very low voltages,
because it would require wires as thick as your arm and they'd still
be too resistive and inductive, so instead you distribute the DC at,
typically, 48 volts. The subsequent conversion to low DC voltages has
to happen via an intermediate AC, but it's a high frequency AC, so it
can be done much more efficiently using ferrite magnetic components,
active rectification, and often resonant mode filters. This high
frequency AC is confined to the internals of a power supply unit, it
never travels over wires or between boxes, thus reducing typical
high-frequency problems such as RFI.

I haven't mentioned battery-backup (i.e. UPSs). They make the system
more complex, but don't change any of the fundamental concerns. Even
on a DC distribution system, the UPS system requires it's own
additional stages of DC->AC->DC conversion, both while charging
(standby) and while discharging (during AC power failure). This is
because battery charging has to have a precisely controlled current
envelope. And batteries don't discharge at the uniform and
well-regulatted voltage that your DC distribution wants. They need
regulators, and switchmode regulators (typically DC->AC->DC) are the
most efficient choice.

Re:AC conversion...

[jsveiga]

We're all concentrating on the electronic, switching power supply stuff.

What about the big power guzzlers in the house: Refrigerators and Air conditioners? Those AC motors suck power directly from 220/110 VAC, and isn't AC better for these cheap induction motors?

This article and the raised-floor article both bad

[sirwired]

The article on raised flooring was an interesting question, but stupid solutions.

That article talked just like some "Intelligent-Design" moron. Just because HE can't figure out how to properly model raised-floor airflow, it must not be possible to do it at all. Wrong. There are any number of companies that will do this for you.

The solution to raised floor airflow is proper modeling of the equipment, vent tiles, and blowers, and relatively unobstructed floor plenum. The solution is NOT air-cooled equipment on bare floor and overhead cable runs. If cooling is still a problem, then use liquid-cooled racks and equipment. (This is where things seem to be going right now.) While overhead cable runs may work fine for some dinky test lab, "real" equipment requires power cables of a size that would quickly fill most overhead runs.

This article proposing DC power is equally stupid.

An enterprise storage box, fully configured that I looked at requires 13,800 kVA of 208V three-phase power (100A inrush current). My mind can barely fathom the completely unbendable copper "wire" that supplying that much juice at 40-ish volts would require.

Telco's switches have a far lower power density than modern servers, and the DC power was made to correct for different problems.

If this guy's ideal data center is overhead cable runs, ceiling blowers, bare floor, and DC power, I'd run away fast.

SirWired

Engineers or Technicians?

[CorporalKlinger]

I think this article might be using the term "engineer" a bit too loosely. I doubt any company would hire an engineer - an actual person with a Professional Engineer's License - to work on these systems. A more appropriate term might be "technician," which usually refers to someone who is trained to repair and work with a single type of technology. Engineers, on the other hand, are usually trained to work with a large variety of technologies and usually work on either (A) Research and development, (B) Manufacturing, or (C) Failure analysis and redesign.

I guess using the term "engineer" sounds better though since it tends to scare the corporate fat-cats away from a technology because of the implied additional cost from hiring an engineer as compared to a technician.

Re:Twice the cables!

[Grishnakh]

Again, it just doesn't make economic sense at all to install all that extra cabling, come up with new standards, etc., just because some people don't like all their wall-warts. Sure, it might be a more elegant solution than wall-warts, but practically it makes no sense at all.

It's different, however, in a data center, where you have hundreds of computers, network switches, etc., each with their own power supply, and also importantly, all in a relatively small confined space. Here, I think (once standards were in place) it might very well make sense to have a standard DC voltage or bus which all the equipment runs on. Because of the sheer number of power supplies involved which would be replaced, and because of the close proximity of all the pieces of equipment, this might very well make economic sense. One large power supply could be made highly reliable and redundant, it would have higher conversion efficiency than many small supplies, and the wiring losses due to lower voltage would be minimal because everything is in the same room, although you'd still probably want to stick with something like 48V rather than 3.3V, 5V, and 12V. Also extremely important, it would make the integration of battery-powered UPS equipment very simple and eliminate any losses there. I believe telcos use 48VDC equipment for these reasons.

Now, for your too-many-gadgets office, it might eventually make sense to standardize on a particular DC voltage, and have a large DC power supply which connects to all your devices in a star or daisy-chain topology. But it would have to be optional, because not everyone has that many devices, or maybe someone wants to install one device by itself somewhere, etc., so this arrangement would need to use a standard connector and standard DC voltage. The problem here, however, is stupid consumers, who might pick a larger DC supply that's not large enough for all their devices, and then blame the manufacturers. When dealing with stupid consumers, it's easier to just include a cheap wall-wart instead of letting them use 3rd-party equipment.