Open Compute Developing Wider Rack Standard

1sockchuck writes "Are you ready for wider servers? The Open Compute Project today shared details on Open Rack, a new standard for hyperscale data centers, which will feature 21-inch server slots, rather than the traditional 19 inches. "We are ditching the 19-inch rack standard," said Facebook's Frank Frankovsky, who said the wider design offered better heat removal and a unified approach to power, including a 12 volt busbar. The Open Compute Project, developed by Facebook to advance open source hardware design, believes an open approach can avoid the mistakes of blade server chassis design."

Comments

Clouseau: The case is solv-ed

[alphatel]

So put the server power supply on the outside, basically.

Re:Clouseau: The case is solv-ed

[a90Tj2P7]

So put the server power supply on the outside, basically.

That summarizes it more frankly than Frank Frankovsky did.

Re:Clouseau: The case is solv-ed

[marcosdumay]

If I understood that correctly, it is not only a matter of putting the power suply outside, but also of making your heat dissipators wider, and space things better inside the servers.

Re:Clouseau: The case is solv-ed

[pinfall]

If I understood that correctly, it is not only a matter of putting the power suply outside, but also of making your heat dissipators wider, and space things better inside the servers.

external power plus common sense? Okay I'll bite. Please describe the miracle achievement.

Re:Clouseau: The case is solv-ed

[NemoinSpace]

planned obsolescence?
I'm getting started on those 19" : 21 " rack adapters now.

Re:Clouseau: The case is solv-ed

[lsatenstein]

Why not go for 30 inches outside. That is the standard size that doorways widths are standardized to. If you allow 4 inches on either inside for cabling, why stop at 21 inches?

 

Re:Clouseau: The case is solv-ed

[RockDoctor]

Why not go for 30 inches outside. That is the standard size that doorways widths are

I thought it was 39.3696in. At least, it is here. Unless you're building from new in which case a significant number of the doors are going to have to be wheelchair-friendly width (I think 59.0544in, but I'm not familiar with that measurement system).

And you forgot to allow at least 3cm for the thickenss of the knuckles that are going to carry the rack through that doorway. Unless you like watching your minions gouge lumps out of their knuckles manhandling stuff through doorways.

(Last time I did decorating, I had to fit 2*29in doors, 1*27in and 1*31in. What is this thing called standardisation again?)

Re:Clouseau: The case is solv-ed

[lsatenstein]

I live in a Metric measure country, however all our appliances and even furniture has to pass through a 30 inch wide opening. Sometimes sofas have to be turned 45 degrees in the length to allow 30 inches overall.

I presumed that 30 inches therefore must be a minimum standard opening in North America.

Re:Clouseau: The case is solv-ed

[RockDoctor]

"standard". I know that word from somewhere ...

Yes, there are standards. Allegedly. But they seem to change. I don't know how frequently they change, but it's often enough that you have to measure every bloody door, find ones of the right sizes, in a consistent style, and that's before fitting the hinges, catches and other bits and pieces.

metric?

[SkunkPussy]

The first mistake is that they are still talking in inches instead of metric units.

Re:metric?

[Anonymous Coward]

Arguing metric vs. imperial units is pretty much the epitome of bikeshedding.
We can do arithmetic nowadays.

Re:metric?

[Christian Smith]

Arguing metric vs. imperial units is pretty much the epitome of bikeshedding.
We can do arithmetic nowadays.

We can do, but people (even rocket scientists!) still get it wrong: http://en.wikipedia.org/wiki/Mars_Climate_Orbiter

Technical issues should use SI units.

Re:metric?

[hobarrera]

While I share the same hatred towards imperial units that most non-US people feel, this doesn't deserve to be downvoted, it's quite true, and GP is just trolling IMO.

Re:metric?

[Anonymous Coward]

http://en.wikipedia.org/wiki/Mars_Climate_Orbiter Technical issues should use SI units.

The Mars Climate Orbiter was a case of someone not labeling their units. The unit system wasn't the problem.
Secondly SI isn't always the best unit of measurement for performing calculations. In plasma physics we use eV in stead of joules for energy because it simplifies our work. In astro physics measuring distances in the SI unit of length, the meter, is impractical. If you are doing anything with relativistic electro magnetism, it's best to do your calculations in cgs because it eliminates useless constants.

Re:metric?

[MetalliQaZ]

Actually, I think deliberations over the color of a bike shed is the epitome of bikeshedding.

Re:metric?

[Anonymous Coward]

Why not use the existing 24" standard, probably because 21" is some metric (1/2 meter) number.

Re:metric?

[Frosty Piss]

21 inches = 0.5334 meters, *not* 1/2 meter.

Not a big difference, but significant for mounting equipment...

Re:metric?

[richardkelleher]

The first mistake is that they are still talking in inches instead of metric units.

This was my thought exactly. If we continue to build new standards around obsolete measurement systems we are just pandering to the Luddites. It is time to move America forward into the 19th century. If we can't engineer for the 21st century, we should turn the creation of standards over to people can.

Re:metric?

[Anonymous Coward]

Arguing metric vs. imperial units is pretty much the epitome of bikeshedding.
We can do arithmetic nowadays.</p></quote>

You have no idea how often this comes up in a particular industry (mine). We deal with critical temperatures in computer rooms (among many other things) and i have had this conversation way too often: "We see the values from your system come through in C, how do we get them in F? My software (of which I am fully in control of the programming therein) is expecting the temperature in F because that is what my user wants. Why don't you convert it to F on your end?" ... "Sir, the operation is val * 1.8 + 32 so if you simply use 1.8 as a multiplier (you have this feature in YOUR software already) and if you use 32 as an offset (you also have this feature already) you can easily show your user F instead of C." ... "Oh, well it would still be easier if you could reprogram your system to deliver the value in F."

Re:metric?

[ChatHuant]

The Mars Climate Orbiter was a case of someone not labeling their units. The unit system wasn't the problem.

But units wouldn't need to be labeled if everybody used the same system. The continued existence of the zombie Imperial system is the root cause of the problem.

Secondly SI isn't always the best unit of measurement for performing calculations. In plasma physics we use eV in stead of joules for energy because it simplifies our work. In astro physics measuring distances in the SI unit of length, the meter, is impractical

Which is why SI has a number of accepted units. You'll note that both the eV and the astronomical unit are there, but not the feet or yards used by Lockheed to send a rocket past Mars.

Re:metric?

[David Chappell]

The first mistake is that they are still talking in inches instead of metric units.

I would imagine that the reason is that 21 inch racks are already a standard. They are widely used in telephony. Introducing a third standard (say 55 cm racks) would likely complicate things for little gain. Better to just call them 5334 mm racks.

This has already been done repeatedly in engineering. When I visit Europe (I am an American) I see things in Imperial units all around me, but most of the time nobody but me knows that they are. 6 mm steel cable? Why not 5 mm? Because 6mm is 1/4 inch. While is their 900 grams of buckwheat in this bag rather than 1000 grams? Because it is a two pound bag filled eight grams short. Railway gage in the former Soviet Union? 1524 mm. Yup, thats five feet. See that 1220x2440 mm sheet of plywood. Yup, 4 by 8 feet.

This by the way is why it is so hard to convince Americans that they should switch the building and engineering trades to the metric system. All our building materials are sized on Imperial units. Converting these dimensions to oddball metric sizes would mean that builders would need to memorizes all sorts of weird dimensions and carry pocket calculators to figure out where the center of something is. There have been many serious attempts to do engineering projects in metric units. First they find out that they can't get materials in even metric dimensions. Then the vendors ask what they mean when they order 914.4 mm doors. Then the builders ask to have the drawings converted into 'the real measurements'. It begins to seem too much like masocism and on the next project they go back to units everybody understands.

Many professions and markets have converted to metric in the US. Medicine mostly has. So has alcohol distribution. So have laboratories. So have our automobiles. But, I don't see any way for the building trades to convert.

The metric system is clearly superior in scientific contexts. There the argument about ease of conversion is powerful. However this argument is meaningless in daily life. The only unit conversion the average person does is feet to inches. You don't need to know how many feet are in a mile in order to know whether you are exceeding the speed limit.

I have found one use for metric units in daily life: increasing ones 'geek cred'. Since expressing dimensions in metric units is a form of elitist obfuscation, it is a great way to be annoying.

Re:metric?

[Anonymous Coward]

But units wouldn't need to be labeled if everybody used the same system. The continued existence of the zombie Imperial system is the root cause of the problem.

Which is why SI has a number of accepted units. You'll note that both the eV and the astronomical unit are there, but not the feet or yards used by Lockheed to send a rocket past Mars.

ALWAYS LABEL UNITS! This was the ONLY cause of the problem with the mars orbiter. If Lockheed used kilometers as their units instead and Nasa assumed meters the probe still would have had problems. If I tell you I have an energy of 5 does that mean 5 eV or J? You don't know if I don't label it.

Re:metric?

[spire3661]

Logically false. You are saying that the existence of a different measuring system is the cause of the human failure to differentiate. It was a human failure, what you are asking for is to dumb it down so humans cant fail in that way anymore. I assure you, humans will find some other way to foul it up, no matter how many rubber bumpers you put on things.

Re:metric?

[jeffmeden]

The first mistake is that they are still talking in inches instead of metric units.

This was my thought exactly. If we continue to build new standards around obsolete measurement systems we are just pandering to the Luddites. It is time to move America forward into the 19th century. If we can't engineer for the 21st century, we should turn the creation of standards over to people can.

Actually most server and rack manufacturers offer specs primarily in cm/mm, and include inches as subtext or a footnote. Upon closer inspection, the Opencompute stuff also offers all drawings this way in mm and in. I suspect they said "this will be 21 inches instead of 19" so it would be more generally consumable, since not everyone finds it as illustrative when they hear "this will be 538mm instead of 482mm". If you can't be bothered to look, then enjoy your time in the 18th century. The rest of us (even if we are in the US) are getting along just fine using the metric system.

Re:metric?

[spire3661]

There is nothing obsolete about the Imperial system. Its a unit of measure like anything else. It may not be as elegant as some other systems, but it works just fine. DO you really need to be able to divide by 10 to feel better about it? The ONLY difference between metric and imperial is that metric is easily understood by idiots.

Re:metric?

[jeffmeden]

I have found one use for metric units in daily life: increasing ones 'geek cred'. Since expressing dimensions in metric units is a form of elitist obfuscation, it is a great way to be annoying.

Hah, too true. I love throwing out "do you mean short ton or long ton?" whenever someone uses the word "ton" as an expletive attributive in conversation. Did I mention that I am a vocabulary pedant too? Anyhow, don't forget about the true pinnacle of intellectual elitism, the completely useless and obscure systems of measurement (a good list is on wikipedia: http://en.wikipedia.org/wiki/List_of_unusual_units_of_measurement) Enjoy!

Re:metric?

[dj245]

But units wouldn't need to be labeled if everybody used the same system. The continued existence of the zombie Imperial system is the root cause of the problem.

No, you should always label your units. My company's (Japanese-based) engineering document system usually lists material stock dimensions in mm. Except for the times that it lists it in meters. And sometimes it is ambiguous where the decimal point should be since that isn't listed (and we often work in 100ths of a mm). There is plenty of confusion because some idiot Japanese designer didn't list his units, and imperial units aren't involved.

Re:metric?

[246o1]

And your argument in turn implies that there's no point in ever trying to be systematically consistent to reduce errors, because .... What? The frequency and severity of human error is going to be constant regardless of the systems people are forced to work within?

People will continue to make mistakes. In some cases, the existence of confusing doubles standards increases the chances of that happening, as well as introducing pointless costs. Measurement is a wonderful example of a natural monopoly, and we should prefer (open) standards.

Logically false. You are saying that the existence of a different measuring system is the cause of the human failure to differentiate. It was a human failure, what you are asking for is to dumb it down so humans cant fail in that way anymore. I assure you, humans will find some other way to foul it up, no matter how many rubber bumpers you put on things.

Re:metric?

[Anonymous Coward]

The ONLY difference between metric and imperial is that metric is easily understood by idiots.

That's not a difference! Imperial is easily understood by idiots too.

Re:metric?

[Jamu]

The inch is a metric unit. As a consequence of the yard being defined as 0.9144 meters, the inch is exactly 25.4 mm. The Open Rack server slots will be 5.334 meters. Maybe they should round down to 5.333... meters.

Re:metric?

[marcosdumay]

Yeah, you should reprogram your system to deliver the value in F. If you already have settings for a multiplier and an offset, why not add measurement unit?

Re:metric?

[atisss]

And on regular human-object level we use metric, which eliminates useless constants and conversions.

Re:metric?

[gstrickler]

But units wouldn't need to be labeled if everybody used the same system. The continued existence of the zombie Imperial system is the root cause of the problem.

Flawed logic. Units always need to be labeled. Here's a metric length for you 5.5

You know it's metric and a length, therefore, it's in meters. But is is nm, mm, cm, m, km? Without labeling, you have no idea what the units are.

Unless you want to eliminate all the si prefixes and list everything as decimal gm, m, l, etc. Of course, that's as absurd proposition, making your statement absurd and incorrect.

Re:metric?

[Archangel Michael]

Imperial units are much easier in some cases, being divisible by a whole bunch of numbers not easily divisible on metric. What happens when you have a one meter (M) whole unit, that you need to divide into 1/3? The nice thing about imperial units, is that typically they are easily divisible by 2, 3, 4 (6, 8, 12) while metric units are not. 1/3 of a meter is how much exactly? 2/3 of a liter?

Metric is nice and elegant, so don't get me wrong, but imperial has its own elegance.

Re:metric?

[GuldKalle]

5 meter rackslots?
Fine way to prove a point

Re:metric?

[Anonymous Coward]

And on regular human-object level we use metric, which eliminates useless constants and conversions.

No it doesn't. Quick how many Millitorr in a Kilopascal? Both are metric units.

http://en.wikipedia.org/wiki/Gaussian_units#Maxwell.27s_equations
See how SI has all these mu_0 and epsilon_0 terms.

Re:metric?

[Grishnakh]

But units wouldn't need to be labeled if everybody used the same system.

This is stupid. Units ALWAYS must be labeled, even with the SI system. How do you know whether you're dealing with mm, cm, m, dm, Mm, etc? Or V, mV, uV, etc.? The SI system is very big on using prefixes to show order of magnitude, and if you don't use unit labels, you won't know what order of magnitude you're dealing with. You can assume it, but we've all seen what happens when you make wrong assumptions.

Re:metric?

[Grishnakh]

You also don't know if that 5 means J or mJ or cJ or kJ or MJ. The SI system is always pushing you to use prefixes so you don't have to write out a lot of zeros.

Re:metric?

[ExploHD]

Better to just call them 5334 mm racks.

I preffer 1337 mm racks.

Re:metric?

[Grishnakh]

It depends on what the system is. If we're talking about a one-wire temperature sensor chip, it may very well be hardcoded to output values in degrees C. It'd be stupid to reengineer that chip to have selectable units, when it's much simpler to move that calculation to a higher level, like in the user interface code. Honestly, I can't think of any place where it'd make sense to do temperature conversions anywhere other than in the very highest levels of the system (right before it's shown to the user). After all, what if you have a customer that really wants to see the temperature in degrees Rankin, or worse some weird custom scale? You want to make all the lower levels of the system change for that? It's much easier to change the highest level for requests like this.

This anecdote sounds like a case of some software person being a complete idiot.

Re:metric?

[Savantissimo]

Not really. cgs or kgs? Bq or Ci? Sv or rem? WTF is a candela when you're measuring any frequency other than 540 x 10^12Hz? What about time and angles and solid angles where the official units are less common than the traditional ones in most uses? Metric still has plenty of weird constants, it isn't natural units. (of which there are various systems, each with some inconveniences) Not to mention economic measurements, a little area the metric system completely ignores.

Converting isn't that hard any more. Use Frink. It will handle practically any unit out there, and if it isn't in http://futureboy.us/frinkdata/units.txt (well worth a read), you can add it.

Re:metric?

[Anonymous Coward]

Torr is not SI unit

Re:metric?

[frank_adrian314159]

1/3 of a meter is how much exactly? 2/3 of a liter?

333.33333... mm and 666.66666... ml respectively. Simple.

How much is 1/3 of an ounce? 1/5 of a gallon? 2/3 of a pound? And is that an avoirdupois, troy, or apothecary's pound?

Methinks you have asked the wrong questions, son.

Re:metric?

[AmiMoJo]

What happens when you have a one meter (M) whole unit, that you need to divide into 1/3?

What happens when you need to divide 1ft into 1/5th? You can cherry-pick random examples of computational inconvenience all day, but metric still has two advantages. Firstly it is easy to plug into a calculator or spreadsheet, making all computation easier. Secondly if you are designing something you can simply use 1200mm instead of 12 inches to get the same divisions in addition to the ease of decimal.

Re:metric?

[dmatos]

1/3 of a metre: 33 1/3cm, or 333 1/3mm. Choose as many significant digits as you like.
2/3 of a litre: 666 2/3 mL.

How much is one fifth of a yard? One tenth of a foot? One eleventh of a hogshead? half a slug?

Re:metric?

[ChatHuant]

This is stupid. Units ALWAYS must be labeled, even with the SI system

Maybe in technical documents, but that's not true in many cases. It all depends on having the proper context. If somebody asks "how old are you", answering "35" is quite unambiguous, because of the shared context. You don't need to explain what units you're using - nobody would suspect you mean 35 weeks or hours (or if you're Han Solo, parsecs). The Lockheed problem was caused by the existence of 2 different contexts, Imperial and metric, close enough to be easily confused. If the Imperial units weren't ever used, there would have been no problem.

How do you know whether you're dealing with mm, cm, m, dm, Mm, etc? Or V, mV, uV, etc.?

Duh, from the context! If you have any knowledge of a particular field you should understand what units there are, to a precision of at least an order of magnitude. If your technical drawing represents a car, nobody with a brain would suspect the length of "5" may mean 5 kilometers or electronVolts. Ambiguity only arises in parts of the world that still use Imperial, because the difference between comparable units isn't large enough for automatic unit checking to kick in - in my example, both 5 meters and 5 yards are acceptable for a car length.

Consider the technical documents from Lockheed - they did not have units, but were accepted and used, and surprise, surprise, nobody created a pocket-sized rocket because they thought the units were actually millimeters (cue Bloody Stupid Johnson). The resulting product did fly, even if not exactly where Lockheed intended :) If the Imperial system hadn't still hung around, the thing would have worked well, labels or no labels.

Re:metric?

[catmistake]

Actually, I think deliberations over the color of a bike shed is the epitome of bikeshedding.

Don't care what color, how long the deliberations last, nor where you put your bikeshed, so as long as its NIMBY.

Re:metric?

[Anonymous Coward]

The parent said Metric not SI. Torr is metric.

Re:metric?

[spire3661]

Assumption. I never stated anything close to what you assumed or thought i implied. The OP stated that the mere EXISTENCE of another unit of measure caused the problem, when the true problem arose from a human incorrectly interpreting the issue. It doesnt matter what we call the units, an idiot is going to fuck it up anyways.

Re:metric?

[Shompol]

However this argument is meaningless in daily life.

The water freezes at 0 degrees C. Yes, the point when you car starts skating on the highway. You could argue that Kelvin scale is more useful in some applications, but what's the point of Fahrenheit?

Re:metric?

[AmiMoJo]

Fortunately building does not require sub-millimetre precision most of the time. In the UK we have 22mm pipes, which is 7/8th of an inch. A pretty random standard pipe diameter which ever way you look at it, but everything seems to fit and nothing leaks. Similarly wood is available in 25mm widths, despite 1" actually being 25.4mm.

I have some friends who work at a builder's merchant. The builders mix units randomly. Most use mm for things like pipe widths, but will also ask for "ten feet of 22mm copper pipe" or a "2.4m half-inch plank". We are the absolute masters of half-arsed metrification but somehow things still get built.

Re:metric?

[aix tom]

In astro physics measuring distances in the SI unit of length, the meter, is impractical.

That always get's me in Sci Fi shows. That they state the distance in "such and such million kilometres" instead of switching to more practical Mega / Gigameters.

After all, nobody is selling "2.9 million kilohertz" processors either, they switch to Gigahertz when talking about them.

Re:metric?

[jmkaza]

At 0 degrees Fahrenheit, to the human body, it becomes frickin' cold out.
At 100 degrees Fahrenheit, to the human body, it becomes frickin' hot out.

Why would humans care whether water is freezing? I have no emotion towards how water feels. When I go outside, I'm more concerned with whether I'm going to be freezing, and at 32F, it's only mildly chilly out.

Re:metric?

[Immerman]

Yes we can, and remembering the conversion factors and using them correctly is a pain in the ass with no real benefit. So given two different different systems of measure which makes more sense to use when creating a new international standard:

(1) A system where all the different units were selected to interact with each other as smoothly as possible, requiring at worst a shift of 3^N decimal places to convert between them, and which every nation on the planet has agreed to adopt as the standard going forward, and which is scaled so that at least the most common legacy units map roughly to the new ones (1L ~= 1 qt, 1m ~=1 yd, etc)

(2) A legacy system created before we understood physics well enough to realize that most different kinds of units are fundamentally interrelated, with no attempt to harmonize even between those that measure the same thing (7.48051948 gallons per cubic foot? Really?). And which is widely use by only one nation whose citizens have been too belligerent to adapt to the new official system.

It's not like it would even cause much trouble with backwards compatibility. Want the racks to fit in the same 24" wide space? Make them either 60dm(~23 5/8") or 61dm(~24 1/32")wide. Likely the only real issue would be the metric-threaded mounting holes being a real nuisance for a company with both sorts of racks in use. How many stripped threads, pissed off techs, and lost man hours would that result in? For that reason alone I could see wanting to stick with the same screws. But still, every aspect of a new standard which can be harmlessly defined in metric terms should be, imperial units can't die fast enough.

Re:metric?

[Grishnakh]

I'm sorry, this is still BS. Yes, with ages, you can omit it because it's pretty obvious, since humans all have similar lifespans and lifecycles. For anything technical, you can't. Go look at ANY engineering drawing; it will always have the units specified. If it doesn't, it's not complete. It doesn't matter if you think you know the correct order of magnitude, because you can be wrong. Units are always specified, usually just in one place so they don't have to be repeated all over the drawing. The fact that Lockheed didn't use any units just shows incompetence and sloppiness. I look at a lot of datasheets in my work, which usually have drawings of electronic parts and also the solder pads for them. They ALWAYS show units, even though anyone competent knows approximately what the physical size of an SOIC-8 package is and could figure out the units pretty easily if they were missing. Even so, to avoid confusion, they're always specified.

Re:metric?

[Surt]

21 inches = 0.5334 meters, *not* 1/2 meter.

Not a big difference, but significant for mounting equipment...

Pfft. Nothing a sledgehammer won't solve.

Re:metric?

[David Chappell]

However this argument is meaningless in daily life.

The water freezes at 0 degrees C. Yes, the point when you car starts skating on the highway. You could argue that Kelvin scale is more useful in some applications, but what's the point of Fahrenheit?

I think you overestimate the convenience of a system wherin water freezes at zero degrees C. After all, users of the Fahrenheit scale know perfectly well at what temperature water freezes and make frequent reference to it.

I have noticed that this is hard for Europeans to understand because they have been taught since childhood to treat "below zero" and "freezing" a synonyms. I met a man from Europe once who found it hard to conceive that these concepts are not necessarily one and the same. The conversation when like this:

He: so, ten degrees below zero Fahrenheit is freezing?

I: Yes

He: and ten degrees above zero Fahrenheit is also freezing?

I: Yes

He: How can this be?

I have given considerable thought to the question of the relative merits of the two systems. The chief advantage of the centigrade system is that it is easy to remember the freezing and boiling points of water. I have been unable to identify other advantages not of the "they (for various values of they) use it, so should you" variety.

The advantages of the Fahrenheit system seem to be:

1) The degrees are smaller, reducing the need for fractions

2) The entire range of non-extreme air temperatures (from the standpoint of the human body) is encompassed pretty snugly in the range zero through 100 degrees. This greatly reduces the need for negative temperatures in daily life. It also makes the system more phycologically satisfying. As the temperature drops to zero degrees, it become dangerous to go outside without taking precautions. As it reaches 100 degrees the same is true. Consider these key temperatures:

0 degrees: so cold an ordinary winter coat is not sufficient for more than a brief time. (Mr. Fahrenheiht used salted ice to produce this temperature.)

32 degrees: water freezes

70 degrees: room temperature

100 degrees: the temperature inside your body and the air temperature are approximately the same. It is getting really hard to throw off body heat.

212 degrees: water boils

In other words, the centigrade system is pegged to temperatures which have great significance to processes involving water. The Fahrenheit system is loosly pegged to temperatures which are important as perceived by the human body in air.

Re:metric?

[David Chappell]

Fortunately building does not require sub-millimetre precision most of the time. In the UK we have 22mm pipes, which is 7/8th of an inch. A pretty random standard pipe diameter which ever way you look at it, but everything seems to fit and nothing leaks. Similarly wood is available in 25mm widths, despite 1" actually being 25.4mm.

I have some friends who work at a builder's merchant. The builders mix units randomly. Most use mm for things like pipe widths, but will also ask for "ten feet of 22mm copper pipe" or a "2.4m half-inch plank". We are the absolute masters of half-arsed metrification but somehow things still get built.

I think this is more common than is generally believed. I know a man newly arrived from Kazakstan who knew all of the pipe diameters in both inches and millimeters. He was aware that the inch sizes were the "official" sizes and the millimeter sizes were the result of conversion.

Re:metric?

[Shompol]

Secondly SI isn't always the best unit of measurement for performing calculations.

While Imperial units are always the worst.

Re:metric?

[Shompol]

Why would humans care whether water is freezing?

When the water is frozen, you

• cannot drink
• cannot swim
• not supposed to lick the flagpole
• cannot get wet
• cannot get attacked by a crocodile
• should not eat the fluffy white powder on the ground when it has yellow tint

It is also essential to set the water temperature below 0 degrees C before adding it to your margarita. So you see, Celsius is more convenient even if you live in exotic countries like Florida!

Re:metric?

[Shompol]

And in C it is:

-60 degrees: keep the dogs inside the igloo

-30 degrees: get your valenki, shuba and treuh

0 degrees: water freezes

20 degrees: room temperature

37 degrees: the temperature inside your body

100 degrees: water boils

115 degrees: a typical summer day in NYC. And this is when I get all confused...

Re:metric?

[Immerman]

True, but unlike in imperial units where factors of 2, 3, and 4 are common, metric prefixes are always talking about a factor of at least 10, and usually 1000. In most contexts that's enough for a quick sanity check to tell you which one is being used. And actually when working on a problem I usually do eliminate all prefixes and decimals using quasi-scientific notation (e.g. if I'm working to 3 significant figures 1.27Mm becomes 127e4m) which keeps things nice and neat and allows derived units to emerge without decimal shifts. The idea being that every tiny detail that I have to pay attention to is both a potential source of careless error, and a draws a sliver of finite attention away from the "big picture". Tables and other intermediate values likely to be compared get quasi-engineering notation(preferring the same exponents to strictly proper form) to allow easier comparison, but remain prefixless.

But yeah, even in that case I'm careful to always list units when doing anything but the most trivial of calculations intermediate calculations, accidentally dropping a single term in a complex calculation can wreak subtle havoc on the results, but will almost always be caught if you make sure your units are combining properly. (I'm actually less concerned about the major havoc, that's usually obvious pretty quickly, it's the subtle stuff that can take forever to hunt down, or slip through the calculation phase unnoticed to cause much more expensive havoc when applied)

Re:metric?

[Anonymous Coward]

shouldn't it be 533.4 mm racks?

Re:metric?

[Anonymous Coward]

mine go to 11

Re:metric?

[Anonymous Coward]

You have obviously never worked in Susskind units before. It is very liberating to be able to set pi=hbar=k=h=1.

Re:metric?

[eugene6]

[[should not eat the fluffy white powder on the ground when it has yellow tint]] There is no situation under which it is advisable to eat the fluffy white powder on the ground when it has a yellow tint.

Re:metric?

[Glendale2x]

I would imagine that the reason is that 21 inch racks are already a standard. They are widely used in telephony. Introducing a third standard (say 55 cm racks) would likely complicate things for little gain. Better to just call them 5334 mm racks.

The telcom rack standard is 23".

Re:metric?

[Anonymous Coward]

I beg to differ. The entrenchment of the illogical Imperial system of measurement is important, because of the time taken (and the risk of errors in calculations) every time you have to convert back and forth between one system and another. We lost a multimillion (multibillion?) dollar space probe because of this, you may recall the spacecraft we sent to Mars that apparently crashed into the planet? Remember? It wasn't supposed to do that.

This is not bikeshedding. Also, from a business perspective, the time wasted doing conversions translates to money wasted and therefore profits lost, all thanks to the refusal of lazy-minded Americans to learn a better system of measures. By the way, think of how much less expensive documentation would be if you didn't have to have everything in TWO systems of measure? Every tool that has to be larger (or has to have smaller writing) so that it can have both metric and the archaic, confusing Imperial "standard" system, costs extra, just to accommodate people failing to LEARN.

The metric system is superior in its simplicity, and versatility; everything is done in powers of ten. You don't have to worry about there being 16 ounces in a pound, and 12 inches in a foot, or 5280 feet in a mile... nor whether or not you're using the right size ounce, for the material you're weighing... (is a Troy Ounce the same as a conventional one? No...)

Consider: how many inches are there in a mile? 5280 x 12... can you do that in your head? While we're using obsolete, stupid systems of measurement, we should also still be using hand-spans, cubits, and stones, as well as drams and so on. Can you picture it? Having your weight on your driver's license being either a single two, or three digit number of kg's, or having to list how many stones, pounds, and ounces you weigh? Or maybe just stones and pounds, since weights tend to vary enough that putting ounces would be kind of pointless... but still, stones and pounds, feet and inches... you could more easily measure weight and height in kg's and cm's...

(Before someone tries to "correct" me on the weight/mass issue, know that I don't care about grams being a measure of mass, that's a bullshit abstraction, since you can't directly measure the mass of an object without observing its WEIGHT, the force of attraction between the object or a reference object and another object, or the Earth, etc. It's really ALL weight, the concept of mass is simply a parenthetical to allow for the fact that we determine weight on Earth by observing the force of attraction between the Earth and the object being "weighed"... the notion of mass being a fundamental, non-derived unit is baloney, but I digress...)

How many meters are there in a kilometer? A thousand. How many millimeters are there in a meter? Again, a thousand. How many millimeters, then, are there in a kilometer? 1000 x 1000. I can do that in my head, and so, probably, can you. Work done in metric is much less prone to errors, even when compared to calculations all in Imperial measures, staying in the same system. Then if you have to go from one system to another, more time is lost, and more opportunity for errors to creep in.

Quick, how many ounces in a ton? Can you do 16 x 2000 in your head? Okay, that's relatively easy, if you factor, and arrive at 16 x 2000 = 2 x 16 x 1000 = 32,000...

Now, how many grams are there in a metric tone? (I still think it should be called a megagram...) There are 1,000,000 grams in a Mg, or "long ton" or tonne, or whatever... It requires almost no thought.

Now how many ounces in 28 gallons?

Metric is easier, it makes sense, and oh, by the way, THE REST OF THE WORLD USES IT. We are the only major holdout remaining using this archaic system of measurements based on the width of a kernel of some kind of grain or another, the length of some dead king's foot, the weight of some particularly massive shit he took once upon a time... or the volume he pissed once after drinking an entire flagon of ale, or mead, or whatever. Why don't we abolish, finally, once and for all, this self-imposed isolation?

It's fucking stupid.

Re:metric?

[Archangel Michael]

10 is divisible by 2 and 5. That's it. Convenient for ..1/2s and 20%
12 is divisible by 2, 3, 4, 6 offering twice as many even divisors as 10. And you think that is less convenient?

Re:metric?

[Anonymous Coward]

In astro physics measuring distances in the SI unit of length, the meter, is impractical.

Not really. The metric system comes with these handy things called prefixes to take care of that problem.

One AU is about 150 Gm, (149.597 871 gigameters), One parsec is about 31 Pm (30.856774879 petameters), One light-year is 9.5 Pm (9.454 254 955 5 petameters). Any area of science that does not use metric is doing so for the same reason people who cling to the archaic Imperial system do, because everything's already in Imperial, and they're lazy. So use gigameters or terameters inside the solar system, and Picameters outside it, or whatever prefix is most appropriate for the distance you're trying to express. ...and no, before anyone asks in response, a petameter is NOT a device for measuring the length of your peter, nor the most convenient unit to use to measure the length or girth of it. For most of you, it's centimeters, for me, of course it's decimeters!

But I digress.

Re:metric?

[Anonymous Coward]

By that same logic, the self-destruct switch on a military piece of hardware should not have any special measures implemented to prevent its accidental use, and it should look like, and be labeled similarly to every other switch or lever on the control panel. Because people will screw up anyway, right? No measures need to be taken when millions or billions of dollars are at stake, to say nothing about people's lives, to reduce the number of ways a catastrophic error can be made?

I assure you, you're argument is the one that is "Logically false".

Re:metric?

[246o1]

It doesn't matter what we call them, but it does matter how many sets of competing standards we have. You are skipping steps in your argument, and your claim that 'a human is going to fuck it up anyways' is just negative bullshit. There are clearly ways to reduce the chance of that - one is to move away from having two competing systems.

Re:metric?

[arkane1234]

so stop using the other competing system.

Re:metric?

[Anonymous Coward]

And SI is the worst system to build stuff in if all your tools are Imperial.

Re:metric?

[arkane1234]

careful, dont confuse him.

Re:metric?

[arkane1234]

false, that's your opinion while the GP was factual.

Re:metric?

[FunkDup]

I think you overestimate the convenience....I met a man from Europe...

So metric sucks because you met a stupid person?

I have given considerable thought to the question of the relative merits of the two systems. The chief advantage of the centigrade [sic] system...

It's "chief advantage" is that is it's consistent across orders of magnitude.

1 cm = 10 mm
1 m = 100 cm
1 Km = 1000 m

1 foot = 12 in
1 yard = 3 ft
1 mile 1760 yd

There is further consistency. 1 litre of water is 1 Kg making many daily tasks such as cooking easier.

Perhaps you should give it considerably more thought.

Re:metric?

[Ryanrule]

Rods to a hogs head. Likes it.

Re:metric?

[Anonymous Coward]

Ah, so you have 12 digits on your extremities as well?

As the GP pointed out, we can cherry pick random examples all day, you missed the point. Even divisibility is not the only thing that counts.

Re:metric?

[Shompol]

GP proposed that in some situations CGS system is superior to SI. I am sorry, but CGS is 100% compatible with SI, as centimetre = 1E-2 m, while gram = 1E-3 kg, etc. This completely misses the point that foot is derived from a foot, which does not facilitate ANY computations because you don't convert feet to megafeet and F=mg does not hold because 1 horsepower != one pound times one foot per second squared.

The biggest problem I see with imperial (besides space probes crashing on Mars) is the HUGE barrier to trade it creates. The only things that can be imported into US have to be manufactured specifically for US. Same can be said for exports. This increases our cost of living and decreases wealth all around just because we are too entrenched to change, while the rest of the world somehow pulled it off.

false, that's your opinion

I used metric for half of my life, and switched to imperial later on, so my opinion should count for something.

Re:metric?

[Anonymous Coward]

I am sorry, but CGS is 100% compatible with SI

http://en.wikipedia.org/wiki/Gaussian_units#Units_for_magnetism

In Gaussian units, unlike SI units, the electric field E and the magnetic field B have the same dimension.

So no. cgs and SI are not 100% compatible. This is unless you mean you can convert between the two. In that case Imperial and SI are 100% compatible.

Re:metric?

[Dracophile]

Sure, but why work with millifurlongs or whateverthefuck it is that imperial unit people work with?

Re:metric?

[ncc74656]

How much is 1/3 of an ounce?

Two teaspoons. So mych for that example.

Re:metric?

[David Chappell]

I think you overestimate the convenience....I met a man from Europe...

So metric sucks because you met a stupid person?

I don't think you read what I wrote very carefully. I did not say "metric sucks". I said that those who were raised on the centigrade (Celsius) system are so used to the idea that zero degrees is the freezing point of water that they see the Fahrenheit system as 'broken'. Both the poster to whom I was responding and my "man from Europe" seemed to see it that way.

When I said "I think you overestimate the convenience..." I meant that in countries where celcius is used schoolchildren must be taught that water freezes at 0 degrees. In countries where Fahrenheiht is used they must be tought that water freezes at 32 degrees. In other words, by this narrow measure (which was chosen by the poster to whom I replied), Celcius does have an advantage, but it is a small one.

I have given considerable thought to the question of the relative merits of the two systems. The chief advantage of the centigrade [sic] system...

Why do you object to my use of the term centigrade? It may be old-fashioned, but does that make its use an error?

It's "chief advantage" is that is it's consistent across orders of magnitude.

1 cm = 10 mm

You are describing the advantages of the metric system and the meter in particular. I was describing the advantages of the centigrade (Celcius) temperature scale over the Fahrenheiht temperature scale.

There is further consistency. 1 litre of water is 1 Kg making many daily tasks such as cooking easier.

Perhaps you should give it considerably more thought.

I intend to. I often make use of the fact that 1 liter of water weights 1 Kg. (In fact, if I had to figure out how much five gallons of water weighed in pounds, I would first convert it to liters and then convert to pounds.) However, I cannot for the life of me think how this might be useful in cooking. Please enlighten me.

Re:metric?

[Anonymous Coward]

But units wouldn't need to be labeled if everybody used the same system.

Go back to your Math books, and leave the Physics to the adults. Once you're ready to accept that units must always be labeled, you can come sit at the table with the grown ups.

Re:metric?

[spire3661]

You are taking my statement to an illogical end. I never said that we shouldnt try to reduce human error. My point is that Imperial isnt going away anytime soon so rather then hoping and wishing, lets recognize the fact that we have differing measuring systems and drill it home. My 'an idiot will fuck it up anyways' means jsut that, someone who is a moron will fuck it up no matter how 'safe' you make it. It doesnt mean 'well we shouldnt try then'.

Re:metric?

[David Chappell]

The biggest problem I see with imperial (besides space probes crashing on Mars) is the HUGE barrier to trade it creates. The only things that can be imported into US have to be manufactured specifically for US. Same can be said for exports. This increases our cost of living and decreases wealth all around just because we are too entrenched to change, while the rest of the world somehow pulled it off.

It certainly creates problems in trade, but it is not true that everything imported into the US is redesigned for Imperial units. In reality the entire industrialized word using using a mixture of metric and Imperial unit goods.

True, the vast majority of bolts sold in stores in the US are in Imperial sizes and I believe the vast majority sold in Europe are in metric sizes, but the sizes of the components in manufactuered goods seldom depend on the target market. For example, Japanese manufacturers have been importing automobiles with metric bolts into the US for decades. Other goods are the same. In turn, the US exports many machines and materials in Imperial sizes to Europe. Machines which pack dry goods into bags seem to be a good example. I have frequently seen one and two pound bags of dry goods in countries which have never used the Imperial pound.

The fact that goods are not redesigned for the US market is also one of the reasons for the generally unfavourable view of the metric system in the US. It is seen as the reason everyone has to have two sets of wrenches, one for the "standard" bolts and the other for the metric bolts.

Re:metric?

[AmiMoJo]

10 is divisible exactly by 4. If your maths skills are that poor I can see why you are having trouble understanding this. Also 12 inches isn't too bad, but how about 3 feet in a yard? Or 1760 yards per mile?

Besides, my point was that division of round numbers isn't everything, and I can assure you that in the rest of the world where we can't divide a metre exactly by 6 (unless you just use the fraction 1/6 of course, or round to 16.6mm) somehow we get by.

Re:metric?

[Anonymous Coward]

I grew up using the Fahrenheiht system, and yes I live in the United States. I also believe it to be broken. Centigrate, Celcius, its makes more sense to me. but you do have to agree that when its 0 degrees out it is cold, there is just a degree of difference on "How" cold it is.

Re:metric?

[toddestan]

The water freezes at 0 degrees C. Yes, the point when you car starts skating on the highway.

That's only if they don't put salt down. If they do, ice won't form until the salt and water mixture reaches... 0 degrees F.

Re:metric?

[FunkDup]

I don't think you read what I wrote very carefully.

Point taken re: your topic, partly. That post only referred to Fahrenheit, however this thread and indeed the sentence Shompol responded to are about the metric system.

Why do you object to my use of the term centigrade?

Celsius is not centigrade because it references Kelvin, not phase changes of water at approximately 1 bar. It's no longer in widespread use for this reason, however upon further thought "centigrade" is correct in the context of your post, and I will use it further below.

I did not say "metric sucks".

Fair enough, "Sucks" is over-stating it, however it's overall a fair summary of what you said. I don't think the European man is a valid argument because I suspect he may be more familiar with the freezing point of Vodka. Although my general point here is that measuring systems should be easy to use by ordinary people, I can't agree with choosing a measuring system on the basis of people that don't understand the difference between scales. I am certain this example is uncommon

I was describing the advantages of the centigrade (Celcius) temperature scale over the Fahrenheiht

I agree that degrees C & F have perhaps the smallest differences of any comparable scales. Fahrenheit is confusing and annoying for metric users because it lacks the salient consistency that is throughout the metric system. To put this in other words, you can easily use the "centigrade" system (in the context of human senses, not measuring instruments) because of said salient consistency, but we can not easily use the Fahrenheit system unless we learn a bunch of seemingly arbitrary numbers, which we don't.

the argument about ease of conversion is powerful. However this argument is meaningless in daily life [quoting across posts here]

Daily life is where it's most powerful. When the measuring system is consistent and intuitive you can do much more complicated things without a calculator. Perhaps the biggest difference of all is that there's a compound effect in usability that is missing from the imperial system.

In fact, if I had to figure out how much five gallons of water weighed in pounds, I would first convert it to liters and then convert to pounds.

'nuff said.

I cannot for the life of me think how this might be useful in cooking

It's increasingly common for recipes to specify mass for a variety of reasons related to accuracy. Measuring mass also means you don't make a whole bunch of measuring cups dirty because you just put one pot on the weighing scales and zero it after each addition.
In the real world you get both volumes and masses, often in the one recipe. When it's easy to convert you don't have to rewrite every recipe that uses volume. I frequently compare recipes from three or more books with their inevitable varied units. Easy conversion means not busting out a spreadsheet to understand the differences.
Over time you come to know the specific gravity of things that aren't water and for which accuracy is not paramount. You know that 83 grams of olive oil is very close to 100 ml.

This is useful in lots of place you might not think of. My high school job was in a vineyard, we often knew the volume of the liquid we had (fuel, wine, chemicals) but the tractors lifting power in kilograms. Life is convenient when you can be confident in your conversion while still sitting on the tractor

Re:metric?

[Anonymous Coward]

Wow. Clearly the GP was talking about "even" divisors, or divisors of the number without a remainder. For 10, they are 2 and 5...

10 is divisible exactly by 4? No different than any other number. By that definition 10 is "exactly" divisible by 3 (3 and one-third, 3 remainder 1).. and 4 (2 and 1/2) or 2 remainder 2.

Any integer is exactly divisible by any other integer... they are all rational numbers. Not every natural is *evenly* divisible by every natural. 10 is not evenly divisible by 4

I'm sorry the education system failed you so badly, fuckwit.

Re:metric?

[David Chappell]

I cannot for the life of me think how this might be useful in cooking

It's increasingly common for recipes to specify mass for a variety of reasons related to accuracy. Measuring mass also means you don't make a whole bunch of measuring cups dirty because you just put one pot on the weighing scales and zero it after each addition.

In the real world you get both volumes and masses, often in the one recipe. When it's easy to convert you don't have to rewrite every recipe that uses volume. I frequently compare recipes from three or more books with their inevitable varied units. Easy conversion means not busting out a spreadsheet to understand the differences.

This is very interesting, so interesting that I might try it. I have heard that kitchen scales are common in some European countries and I have heard of European cook books which give quantities of dry and liquid ingredients by weight, though I don't think I have ever seen such a book, at least not one that was written in English. I believe kitchen scales are available in the US, though they can be hard to find. They are mainly used by diabetics to measure portions.

I believe my contribution to this discussion has been generally misunderstood. I am not against either measurement system. I switch systems depending on what I am doing and what those who are doing it with me understand.

I got involved in this discussion because I saw that the advocates of the Metric (deliberately capitalized) system were repeatedly advancing arguments which I know Americans do not find convincing. The advocates kept emphasizing the technical superiority of the Metric system and expressing the belief that those who have not yet adopted it must have failed to appreciate its technical superiority. (Many expressed this belief far less kindly than I have here.)

I believe that many if not most Americans are already convinced of the technical superiority of the Metric system, at least in theory. The problem is that for them technical superiority is not the deciding factor. For them the Metric system is a new and superior technology which they are not comfortable using and which they know is very expensive. (Metricication has many of the same costs as computerization.)

Even dramatically superior technologies consistently face huge cultural and organizational hurdles before they are adopted. The telephone, the personal computer, and the Internet all did. (See "Father Lets in the Telephone" from "Life With Father" http://gutenberg.net.au/ebooks06/0608341h.html#c20) During the period between when the public knows about the technology and accepts it into their lives the advocates have a frustrating time of it. In the 1980's I participated in conversations which went something like this:

Son: Dad, we should get a home computer.

Father: Why? Do you want to be a mathemetician?

Son: Computers aren't just for mathemeticians. You could use it instead of a typewriter. A word processor is much better.

Father: But if I understood you correctly last time we discussed this the required equipment would cost $7000 and take up all of the room on my desk. A typewriter costs only $100 and I can pick it up and put it in the closet when I am not using it.

Son: But it is easier to correct mistakes on a word processor

Father: Perhaps in theory, but in order to gain that small convenience I would have to accept multiple huge inconveniences and pay enough money to buy a car.
Son: Yes, dad, but a computer is a multi-purpose machine. You could use it for lots of other things. You could balance your checkbook too.

Father: All I need to balance my checkbook is a $50 calculator.

Here my father was not denying the technical superiority of the personal computer. He simple did not yet want to do any of the things which it made possible.

I see the Metric system advocates making the same mistakes I made in conve

Wow. Awesome advances in technology!

[Anonymous Coward]

They're changing the size of the hardware racks? HOLY SHIT, MAN. That's some crazy shit right there!

What's next? Changing the lightbulbs in the server room from fluorescent to LED?

So glad Slashdot is right on top of these great advancements in computing technology.

Re:Wow. Awesome advances in technology!

[xeno314]

Well see, we needed to boost revenues in rack hardware. With this, we can win either way! People buying new servers will need new racks, so $$$. People buying more racks who need to use some new and some normal equipment will have to buy adapters, so $$$.

I don't know whether this is really a net win on space vs. adding height to servers that need it in the current form factor, but hey, time to change things up, I suppose.

Re:Wow. Awesome advances in technology!

[networkBoy]

As soneone responsible for machine infrastructure, this is a big deal.
21" racks means that I have to change the layout of my lab. Currently we have the mandatory 42" aisle (or whatever ADA requires) running the length of the lab, with banks of racks pedistileing (sp?) out from the wall with power and data. If all the racks were 2" wider I'd lose a rack in each row, so unless these machines improve density... it's a no sell for me.

Re:Wow. Awesome advances in technology!

[Anonymous Coward]

I hope you enjoy never upgrading your hardware.

Re:Wow. Awesome advances in technology!

[Doc Hopper]

That's covered in the article. 21" racks within the current 24" standard rack size. By eliminating cable mass in favor of bus-bars, you gain the two inches "free" in your rack footprint. If the rack's designed right, you could even keep the same seismic bracing.

spinal tap hardware

[Hognoxious]

These are better. They're 2 batter.

I don't see the point, to be honest. And aren't things supposed to get smaller as technology advances?

Re:spinal tap hardware

[lister king of smeg]

not necessarily, they will get built with higher computational density but the size will remain constant because the workload will increess enough to match moores law

Re:spinal tap hardware

[Grishnakh]

I haven't read TFA (this is Slashdot; we don't do that sort of thing), but maybe they're proposing making the units thinner too. If "1U" is thinner with the wider standard than with the 19" standard, then you should be able to fit more servers into a rack, making up for the increased width.

Re:spinal tap hardware

[megabeck42]

Nope, they increase a "U" to be 48mm from 44.45mm. This is now called an OU. So, 1OU = 48mm or an increase of 8% compared to a regular U. They claim that this 3.55mm increase will "increases airflow, improving air economization; it also allows for better for cable and thermal management and efficient use of space." Personally, I question wether the increase in airflow, cable management, and efficient use of space will be significant. I'd be very keen to see a good example of how these new 48mm rack units will improve cable management.

Also, the bus bars depicted in the photos appear to be incredibly vulnerable to accidental short circuiting.

Re:spinal tap hardware

[Grishnakh]

This sounds like a bunch of marketing BS to me (remember, "marketing" == "lying").

How on earth does making a "U" larger improve cable management? I don't see how it'd make one bit of difference really, except for making it so you can't put as many servers into one rack, and decreasing the total number of cables in that rack, but that's not exactly an improvement. You can do the same now by simply leaving some slots empty or loading the rack half-full, but what's the point in that, since it's just wasting space?

The one (very small) improvement I can see is that with a taller 1U chassis, you can use slightly larger fans inside, and larger fans are more efficient than small ones. Those tiny fans they use in 1U servers are loud and noisy and don't move that much air. But an 8% increase in height isn't going to make much difference here.

2" shim market

[RichMan]

I detect a few years of market window for rails with 1" shims attached (19+2x 1" = 21) to allow old servers to wedge into the new racks.

Re:2" shim market

[istartedi]

No way man. Spring-loaded adjustable shims. I'm off to the patent office. Just kidding; but I know somebody else probably IS off to the PO and isn't kidding.

Re:2" shim market

[MarkGriz]

Or maybe Facebook becomes the new Myspace before this gains any traction, and we can avoid this altoghether.

Re:2" shim market

[gladish]

So you're saying the business to be in is reselling chinese made server rack shims.

Re:2" shim market

[nschubach]

The problem with 1" shims is that the copper power bars are still going to be in the way of the long server chassis. The pictures I saw had the copper bars running up the middle of the rack limiting your usable depth by quite a bit. (page 5)

Re:2" shim market

[David Chappell]

No way man. Spring-loaded adjustable shims. I'm off to the patent office. Just kidding; but I know somebody else probably IS off to the PO and isn't kidding.

I don't see why they need to be adjustable. And you shouldn't be able to patent them because 21 inches is already as standard rack size. In fact, I think I have seen these shims somewhere.

Re:2" shim market

[David Chappell]

Looks like I remembered incorrectly. Telecom racks are actually 23 inches wide.

Re:2" shim market

[istartedi]

The adjustability is there to accomodate all the other proprietary widths that will come down the pike. Once you open that box, and start throwing in metric vs. standard, there's the potential for all kinds of different racks to end up in your cooler. That's especially true if you've got somebody who likes to scoop bargains up from random places. Pennywise and pound foolish to have mixed equipment? You bet; but it happens.

Besides. There's no good reason for a lot of product features. That doesn't stop them from being introduced. The adjustable shim will come with a coupon for elephant repellant.

Re:2" shim market

[Anonymous Coward]

We've used 23" racks for years -- this is hardly a new idea. It is a bit of a pain attaching the aluminum tabs to 19" gear, but the space makes running cables and manipulating systems vastly easier.

"awesome"

[Anonymous Coward]

More and more standards. That's what everyone and the world always needs.

Not.

Re:"awesome"

[ArhcAngel]

Obligatory XKCD reference.

Re:"awesome"

[Anonymous Coward]

More importantly, breaking on of the VERY FEW damn standards that almost everyone seems to have complied too!

Seriously, in a world where no one can agree to use the same anything.. this was a small bastion of hope. The fuck is wrong with these guys!

Re:"awesome"

[PerfectionLost]

Everyone except the telcos.

Re:"awesome"

[ckaminski]

Which was really "Just AT&T" until 1984.

Re:"awesome"

[netwarerip]

Telcos?
And I thought this was a technology website :(

Re:"awesome"

[Anonymous Coward]

Just because something is old doesnt mean it is good.

Re:"awesome"

[PerfectionLost]

Telcos?

And I thought this was a technology website :(

Because clearly telephones work by rubbing to sticks together to generate smoke, which is then run into tubing and out the other end where it is interpreted by tiny monkeys that impersonate your caller's voice in side of your telephone. You don't need technology for that, just a lot of tiny monkeys.

Server width is changing Rack Width isn't

[darthcamaro]

There is a big distinction that you need to understand here, the Width of the Server chassis is changing, not the width of the rack itself. The outer dimension of the Server Rack is staying at 24 inches. The REAL problem was a bogus amount of extra cruft in the rack design that is going to be eliminated to make way for the wider servers.

Re:Server width is changing Rack Width isn't

[afranke]

Yay, so that means that I can still use my lack rack. :)

Re:Server width is changing Rack Width isn't

[Anonymous Coward]

Cruft like structural integrity to hold multiple 200 lb servers (i.e. Sun Thumpers)?

Re:Server width is changing Rack Width isn't

[IGnatius T Foobar]

That "extra cruft in rack design" is where your cables go. I have an 18,000 sqft data center over here and I can tell you from experience that what you call "cruft" isn't nearly enough space for all of the cables once things start getting dense. We are actually considering 23" (telecom standard) racks with 19" rails in them for cabinets that aggregate the networking gear, just for this reason.

But oh no, far be it from Facebook to actually work with the industry. They screwed up the Internet and now they're going to screw up your data center.

Re:Server width is changing Rack Width isn't

[thereitis]

As technology advances, the time it takes to call a predecessor's design 'stupid' shrinks.

Re:Server width is changing Rack Width isn't

[Anonymous Coward]

Facebook is in the data center industry. They already have more experience and larger centers than you have ever dealt with sweetheart.

Re:Server width is changing Rack Width isn't

[Anonymous Coward]

Facebook is in the datacenter industry. They have more experience and much larger centers than you have ever dealt with sweetheart.

Re:Server width is changing Rack Width isn't

[Anonymous Coward]

Have you considered cable consolidation with infiniband with Virtual Protocol Interconnect (VPI)? That kind of architecture migration wouldn't be minor, but it could certainly cut back on cable and infrastructure jumble....

Re:Server width is changing Rack Width isn't

[Wesley Felter]

IIRC Thumper had 48 drives in 4U but Facebook's new hinged storage server has 15 drives per U, so it's even denser.

Re:Server width is changing Rack Width isn't

[Immerman]

That "extra cruft in rack design" is where your cables go.

Which will no longer be needed because you'll be using bus bars instead. Network cabling can be routed through purpose-built channels for which there will now be considerably more space.

It would seem to me that the next step would be to incorporate some sort of standardized ventilation ducting so that cooling could be managed on a rack level as well, allowing for the use of larger, more efficient, more redundant, and *much* quieter ventilation fans. I'd likely design it as a forced air system with a moderately pressurized cold duct which would provide a steady stream of clean, cold air to a shuttered port on each server. Each server could then simply open its shutter as wide as necessary to get sufficient air flow, possibly using the exact same control circuit as would drive the fans if you opted for a non-ducted configuration. An optional chimney could then carry away the heated air in a controlled manner, or it could simply be vented in to the room as normal. And similarly to the bus pars, the cold duct could be "powered" by a either a rack-based cooling system that pressurizes ambient air, or an adapter that ties directly into the HVAC ducting.

Obviously you could instead generate vacuum in the chimney instead, but positive pressure has numerous advantages such as increasing the heat-transfer capacity of the air (slightly, but every little bit helps) and keeping dust out of the servers (why struggle to maintain a server "clean room" when you can just ensure that the cold air supply remains clean) .

Ideally I'd shoot for a form-factor that would be suitable for centralized liquid-cooling as well, allowing the rack to be used with the cooling method of choice without it having a major impact on the layout.

Re:Server width is changing Rack Width isn't

[Anonymous Coward]

18000 sqft isn't exactly large by datacenter standard (unless you're enterprise or just telco). By Facebook's standards? It's minuscule. Get over your silly problems of cabling.

Re:Server width is changing Rack Width isn't

[megabeck42]

The facebook hinged storage server must be using their new 21" rack because they (from images) appear to have an arranged the drives in three rows of 5 drives. The 3.5" drive formfactor is 4" wide, meaning that the enclosure must be at least 20" wide to accomodate five drives per row. Also, using their new rack concept, their servers don't include and AC power supply. So, it's not exactly as space efficient when you factor in the 2U power supply at the bottom. With one PS and one 30drive facebook server, you're at 30drives for 4U or an efficiency of 7.5drives/U. One PS and two 30drive servers, you're at 60drives for 6U and an efficiency of 10drives/U. One PS and three 30drive servers, 90 drives on 8U and an efficiency of 11.25 drives/U. At four servers on one PS unit, you've got 120 drives occupying 10U for an efficiency of 12 drives/U. So, once you have four servers together with the associated PS, you finally reach the efficiency of a thumper.

The thumper (Sun x4500 and x4540) had 48 3.5" HDD's, 2 (x4500) or 3 (x4540) 800W/1600W (110VAC or 220VAC) power supplies and an adorable, itty-bitty dual opteron server. 48 drives occupying 4U is an efficiency of 12drives/U.

To be fair, while the 4server, 1power supply configuration only equals the storage density of the thumper, it has better server/cpu/nic density.

As an aside, the full rack setups appear to have three power supply units. Assuming/guessing 42U per rack with 6U devoted to PS, it leaves 36U divided into three bays of 12U. So, with a PS and five facebook hingy servers occupying 10U and sporting 150 drives at 15 drives/U, you finally outdo the thumper.

To be clear, I do believe that there are benefits to the proposed new rack size, but I don't think it's a clear improvement. Personally, I think the thumper design was brilliant. The only purpose of this reply was to point out that it's not as simple as 15 drives per U.

On a separate note, the ability to fit 5 drives side-by-side in a 21" rack is the best justification I've seen, so far, for widening to 21".

Wow, I prefix too many of my comments with insecure clauses devoid of information and only serving to indirectly apologize to the reader for supplying information I think is important for them to understand despite my worry that I'm trying their patience. If you read all of this reply including even this sudden instrospective insight to my character; then, thank you. I'm flattered.

Re:Server width is changing Rack Width isn't

[Anonymous Coward]

Indeed, increasing density makes the 23" racks appealing, and conserves the heck out of the skin on my knuckles.

already have 23"

[Anonymous Coward]

There's already a standard for 23" racks widely used in telecom. So now we have to deal with 19, 21, and 23 options? Great.

Re:already have 23"

[atisss]

Next - 19 3/4" racks coming to a data centre next to You.

Re:already have 23"

[catmistake]

There's already a standard for 23" racks widely used in telecom. So now we have to deal with 19, 21, and 23 options? Great.

Don't panic. The only rack standards a healthy male ever needs to concern themselves with, especially in Spring and Summer, involves the comprehension and visualization of cup-sizes, A through GG.

Re:already have 23"

[WebCowboy]

There's already a standard for 23" racks widely used in telecom. So now we have to deal with 19, 21, and 23 options? Great.

Perhaps that isn't as great a problem as it sounds. If the overall footprint of all those rack formats is 24" it stands to reason that there would be racks out there where the mounts for the vertical posts have mounting holes to accomodate 19 and 23, and to support 21 would be a matter of putting one more hole in each bracket.

That said, maybe it would've just made sense to say data centres should adopt 23" width and specify different power rail designs to match.

Why invent a new standard?

[Anonymous Coward]

Why invent yet another new standard when there's a perfectly good one already in use, with lots of inventory in place?

Telecom has used 23" racks for years. There are standard adapters already available to mount 19" hardware in a 23" rack. 23" racks are already available in the marketplace.

Further, why 12Vdc? Telecom has been using DC plant for decades and there is a ton of existing 48Vdc equipment on the market. There area existing cabling standards for all this as well.

Just seems like an attempt to fracture the market and create an opportunity to solve a problem that's already been solved.

Re:Why invent a new standard?

[OzPeter]

Why invent yet another new standard when there's a perfectly good one already in use, with lots of inventory in place?

FB IPO .. FB IPO .. FB IPO .. what other reason do you need??!?!?!

Re:Why invent a new standard?

[Tekfactory]

Doesn't Google use 12Vdc straight to their Motherboards, they have 12Vdc batteries for each server, and everything that needs 5Vdc gets power from the motherboard and not a power supply.

I would assume FB is trying to use the same equipment, I don't know if 48Vdc is as prominent on the server side.

Re:Why invent a new standard?

[hjf]

Also, 12VDC? REALLY? so you have like what, 12KW (being VERY conservative) worth of gear in a 42U rack? that's 1000A right there! How the hell do you keep power loss low with 1000A?

Re:Why invent a new standard?

[mpoulton]

Mod this AC up! These problems have already been solved - decades ago. Sometimes it's better to look for solutions to your problem in another industry than to engineer it yourself. In fact, it's almost always better.

Re:Why invent a new standard?

[Anonymous Coward]

12V DC is a great idea. Eliminate the power supplies inside each individual server and run the entire data center off one massive PSU.

It's ridiculous to take AC power, convert to DC to charge and run the UPS, then convert back to AC, only to have individual servers convert AC back to DC.

However, I think any new standards should also take a serious look at full blown water cooling solutions for main heat sources. Concentrate most of the heat into liquid and reduce the amount of air conditioning required. Easier to cool water than it is to cool air.

Re:Why invent a new standard?

[avandesande]

... the AC post should be nominated for one of the best ever

Re:Why invent a new standard?

[networkBoy]

As others have stated, this is remarkably inefficient. There is a reason Edison lost the power distribution battle.
I think you may be able to run one PSU per rack, it's in the middle, with bus bars top to bottom, but beyond that you're looking at I^2*R losses that run over what you lose in the AC- DC - AC DC cycle of mains - UPS - PSU - MB
-nB

Re:Why invent a new standard?

[couchslug]

Fracture markets for vendor-lock win!

Re:Why invent a new standard?

[Anonymous Coward]

12V DC is a great idea. Eliminate the power supplies inside each individual server and run the entire data center off one massive PSU.

No, it's a stupid idea thought up by people who are ignorant in matters of distributing electricity. When you reduce voltage you increase amperage, there's no free lunch. Running bus bars around rated in the thousands of amps and up to distribute 12VDC is ridiculously stupid.

Re:Why invent a new standard?

[Glendale2x]

Because they called this one "open" and used a lot of green colors to imply its "greenness".

Re:Why invent a new standard?

[Anonymous Coward]

Very good point on the 48Vdc. It's also the standard voltage in 802.3af Power-over-Ethernet systems. You can push the higher voltage at a lower current.

Re:Why invent a new standard?

[Anonymous Coward]

Because this IS NOT changing the actual exterior rack width, it's deleting the 'dead space' older racks left for mounting brackets, side-routing cables, etc; most of which modern servers already eschew so it's just "dead air" space between racks that is outside of the airflow path so it's stale air.

Telecom racks are physically wider: 32" floor spacing rack-to-rack w/ 23" internal rail spacing. Data racks are 24" spacing w/ 19" internal rail spacing, this 'new standard' is purely to change data racks to 24" spacing w/ 21" internal rail spacing; in effect ditch some of the side-gaps entirely so servers can utilize the space for components.

Re:Why invent a new standard?

[Anonymous Coward]

If you think it's a great idea then fine, you go stick your hand in the rack with the 12v bus bar with 1000A running through it. That's the kind of excitement I'd rather not have, personally.

Re:Why invent a new standard?

[afidel]

Close, Each bus bar pair can support up to 500A, which can be increased by adding more copper. Each bus bar pair can support up to 6KW, while power can be increased by using more copper. The bus bars can be adjusted for higher current, depending on the desired power per column, as they are interchangeable. Normally the Open Rack has three bus bar pairs installed, but it is also configurable with two or one bus bar pairs in the power zone: link

Re:Why invent a new standard?

[rickb928]

It's not about the resistive losses, mate.

First, eliminating power supplies at each server eliminates multiple point heat sources. One honking rectifier stack somewhere does the job. Cool that separately, and maybe even more efficient conversions, though the jury is still out on that for me. Honking-er bus bars will not be as efficient as AC distribution, but the losses will be tolerable. Remember the stack? Getting that cooled outside the server room is useful.

Using 48VDC makes a lot of sense just because there is a pile of telco equipment made to do that, and well understood. It will have to be regulated at the server level anyways, and the current goes down also.

But then again, it's not about being compatible, nor even evolutionary. This is a revolution, and if the old guard can't be enslaved, it must be killed off. So telco standards are out of the question. Besides, those extra 2 inches.

Re:Why invent a new standard?

[afidel]

Actually they are doing three power zones per cabinet fed by 277V AC from utility and 48V DC for backup power which then powers the 12V bus bars.

Re:Why invent a new standard?

[Bill_the_Engineer]

You need the higher voltage supply to make up for voltage drop due to resistance found in the power cable themselves. The longer the cable the larger the resistance and the more pronounced the voltage drop. You could compensate by increasing amperage (to make up for lost power) and using heavier gauge cables (to reduce the internal resistance and handle the increased amperage), but the accepted practice is to distribute a higher voltage to the equipment and use a DC-DC converter in the device or immediately at its power input to deliver the required voltage. This is because it's generally safer and cheaper (material costs, weight) to distribute higher voltage than higher current.

This is why telcos use 48Vdc and the aircraft industry (MIL-STD-704F) use 24 Vdc.

Re:Why invent a new standard?

[Vairon]

Computer motherboards don't run on -48V DC. They run on +12V DC. Power supplies normally take the 110-220V AC power and convert it to 12V DC for the motherboard and 5V DC for peripherals but this causes a loss of efficiency. This Open Rack infrastructure puts the power supply which they're calling a 12V power bar on the rack and supplies the computer with exactly the 12V it needs. In the article they claim they got 50% power savings over a traditional power supply/computer design. While there may be a lot of telecom gear for -48V DC that is not what a computer motherboard needs. It would still require a power supply to convert -48V DC to +12V DC if they had gone that route. This would have increased cost and decreased power efficiency.

Re:Why invent a new standard?

[SuricouRaven]

This is in contrast with the standard ATX supply which runs, at a minimum, rails for 5V, 12V, 3.3V and -12V. The latter is really a leftover from powering old RS232 ports, but the standard still requires it. Even then the pins are so small that many pins in parallel are required, thus the sight of a ridiculous twenty-six-cable tangle winding it's way through your case like like a nest of intertwined worms.

Re:Why invent a new standard?

[Anonymous Coward]

It won't shock you unless you have studs piecing or open wounds that allows electricity to conduct more freely.

On the other hand if you drop a screw driver or any bare piece of metal that shorts it together, you'll have a major case of high current getting the metal to go red hot or even vaporize.

I wonder how the heck they are doing the power distribution inside the rack itself. i.e. what kind of insane size power connectors do they use? Car jumper cables to each server card isn't exactly usable.

Seriously guys. Last time I worked with -48 equipment 5 years ago, they had 97% efficiency unregulated PUPS (1/2 bricks) to down convert -48V to 12V. Just exactly what home made crappy converters did they used for their calculations for the 50% gain in efficiency?

Re:Why invent a new standard?

[petermgreen]

Telecom has used 23" racks for years. There are standard adapters already available to mount 19" hardware in a 23" rack. 23" racks are already available in the marketplace.

AIUI they don't want to increase the overall size of the rack so they can fit into standard datacenter floorplans, presumablly this limits the ammount they can widen the server space in the rack without widening the rack itself.

Telecom has been using DC plant for decades and there is a ton of existing 48Vdc equipment on the market.

48V DC was well suited to the needs of telcos at the time who originally used it to directly power POTs lines, that doesn't mean it's appropriate for powering computers.

To prevent currents flowing through metalwork causing ground potential shifts and to limit damage in the event of faults you generally want your computing gear to be isolated from your building power distribution system. This generally implies having an isolating PSU (whether AC-DC or DC-DC) between your building power distribution and your motherboards.

Afaict this 12V system is about connecting multiple servers in a rack to a common rack level PSU, not about changing building power distribution standards. Presumablly 12V was chosen beause it is becoming the standard for powering motherboards (5V and 3.3V are mostly legacy supplies nowadays)

Re:Why invent a new standard?

[tlhIngan]

Close, Each bus bar pair can support up to 500A, which can be increased by adding more copper. Each bus bar pair can support up to 6KW, while power can be increased by using more copper. The bus bars can be adjusted for higher current, depending on the desired power per column, as they are interchangeable. Normally the Open Rack has three bus bar pairs installed, but it is also configurable with two or one bus bar pairs in the power zone: link

That bus bar had better be pretty damn big cross-sectionally. 500A is a LOT of current. And if there's three, that's 750A. You better put that power supply at the bottom of the rack, because the IIR losses are huge (they grow with the square of the current - pass twice as much current, and the loss goes up 4 times).

Pretty much the only way to have racks of it is to make the roof and floor out of solid metal and weld the racks to both sides (IIR losses are huge at connection points). Hell, that's how welding works - high currents, low voltage. Any bit of dirt between the rack's bus bars and the power cable will weld them together over time, if not cause a fire due to heat.

Hell, the voltage drops are pretty big as well. 500A, if the connection points add 0.01 ohms, you're looking at 5V. (And using IIR, that's a good 250W of power lost). Get it under 0.001 ohms and you're looking at 25W dissipation in the connectors and still half a volt of drop, which is nearly 5% voltage drop.

There's a reason why telecoms use 48V, and why electrical towers run at hundreds of kilovolts. It's the best mix of voltage conversions and acceptable power losses. Or why datacenters have other voltages like 208V or 240V readily available - lower the current and the cables can be made much smaller.

Heck, even *CARS* are considering switching to a 48V system to be able to power everything.

Re:Why invent a new standard?

[Bill_the_Engineer]

This is a revolution, and if the old guard can't be enslaved, it must be killed off.

Touching the 12Vdc @ 500A copper bar bus will kill off some of the new guard too.

Re:Why invent a new standard?

[nsaspook]

Using 12VDC on any type of system that uses more than a few hundred watts is insane. Only a Best Buy boombox tech would think it was a good idea.

Re:Why invent a new standard?

[afidel]

A power zone is no more than 1/3r'd of a rack (and can be as little as 1/9th of a rack but the spec still calls for the bus bars to cover the entire 1/3rd of the rack). Oh, and as far as 500A being a lot of current, sure but datacenters deal with large currents all the time, for instance the inputs to my UPS's are 300A at 480V and they're small by datacenter standards.

Re:Why invent a new standard?

[afidel]

These people aren't idiots, they've got very intelligent engineers and they are trying to solve a very specific problem which is how to minimize the cost per transaction for web scale platforms. When you've got a million servers (Google's been there for some time and Facebook is probably well on their way) a few percentage points in efficiency can be very real money, and if your design doesn't work for everyone that's ok because manufacturers will be happy to do a custom run of 50k units for your new small test datacenter.

Re:Why invent a new standard?

[RulerOf]

the sight of a ridiculous twenty-six-cable tangle winding it's way through your case like like a nest of intertwined worms.

The mental image on that one is priceless :D

Thank you, sir!

Re:Why invent a new standard?

[rev0lt]

Oh, and as far as 500A being a lot of current, sure but datacenters deal with large currents all the time

Usually not using DC, and not on a computer-tech acessible zone. Not only 500A can easily melt/vaporize metal (electric welders usually are on the 100A-200A range), but as the parent pointed out, the voltage drop is significant. And there is another motive why this is not very smart - ticker conductors are heavier and more expensive. Considering the increasing cost of copper, this will translate on an substantial increase in cost per rack.

Re:Why invent a new standard?

[rev0lt]

Using 12VDC on any type of system that uses more than a few hundred watts is insane

Like in a car?

Re:Why invent a new standard?

[afidel]

And that increased capital cost will be completely and utterly dominated by the increased efficiency in power useage. If you think these guys haven't analyzed the numbers every which way and run a pilot datacenter to prove out their numbers before releasing this stuff you're delusional. These are not stupid people and they know exactly the problem they are trying to solve. It might not be the same problem as everyone else in the computer sector is trying to solve, but it is a problem quite a few very large web companies are trying to solve (ie this is a solution for Google, Microsoft, Facebook, Amazon, and a few other very large companies with hundreds of thousands to over a million servers).

Re:Why invent a new standard?

[Anonymous Coward]

[...]Heck, even *CARS* are considering switching to a 48V system to be able to power everything.

42, actually, and that was the nominal charging voltage, for a 36-V battery. But that seems to be dead. http://en.wikipedia.org/wiki/42-volt_electrical_system

Re:Why invent a new standard?

[ZeroSumHappiness]

If you search you can find a real image so you don't have to make a mental one. It's quite fascinating, actually.

Re:Why invent a new standard?

[marcosdumay]

Further, why 12Vdc?

Because that is what the current servers actualy use. Why put a 48V bus on the rack if you'll still need to convert it to 12V?

Re:Why invent a new standard?

[marcosdumay]

Touching the 12Vdc @ 500A copper bar bus will kill off some of the new guard too.

How? Are you trying to tell me that those 500A will run inside a person's body when you apply 12V of potential diference on it? You mean a person's body has 24m Ohm of total resistance?

The 48V bus would kill much more people, even if less current flowed through it. A person only needs a few mA to die.

Re:Why invent a new standard?

[sjames]

Copper is expensive and heavy. I would rather see 48v supplied and lighter bus bars.

Switching power supplies can be quite simple and efficient if they don't need PFC or boost capability. That suggests DC input above the maximum output voltage.

Re:Why invent a new standard?

[nsaspook]

Like in a car?

Yes, like a car. Other than the starter battery for a few seconds the rest of the cars uses under about 400W (with 10%+ lost in wiring) even with the headlights,fans running, and much less without. Nobody in their right mind would use a 12VDC buss for a 3+kW power distribution system. The feeder fusing requirements would be so high that a dropped tool would melt before the protection opened the circuit.

Re:Why invent a new standard?

[Grishnakh]

They should just dump the -12V requirement. They've had chips available for years that can supply +/-12V for RS232 ports; I believe the MAX232 from Maxim is one of them: you connect it to 5V data lines from your microcontroller, give it 5V for power, add in four capacitors, and it outputs +/-12V RS232 signals. I was using this chip back in the late 90s for embedded equipment that needed RS232 connectivity so we wouldn't need a power supply with -12V.

They should also dump the 3.3V requirement. AFAIK, that's only used for SATA hard drives, and even there most don't use it because they couldn't count on people actually having 3.3V-compliant power supplies, as many users were installing the (then-new) SATA drives into older computers and using Molex-to-SATA power adapters. 3.3V has a huge drop over any distance, so it makes much more sense to just use a DC-to-DC convertor at the point of use to convert 12V to 3.3V. Now that even CPUs want 12V instead of 5V, and high-end video cards have their own 12V connectors, it seems to me they should just dump all the other requirements altogether, and make a PC power supply only produce +12V, and have the motherboard and drives use their own DC-to-DC converters to produce whatever voltages they require.

Re:Why invent a new standard?

[Grishnakh]

Telecoms use 48V (actually -48V) because they've used it for many decades, long before silicon ICs were invented.

That said, I really wonder how the losses would compare for using 48V and using DC-to-DC converters to produce 12V on each server. Also, how would the cost compare? Copper bus bars are expensive, and copper is getting more expensive all the time; before long, people will be thinking of using silver instead (exaggerating, but only slightly).

Re:Why invent a new standard?

[DigitalSorceress]

Yup, that's what I was thinking - I used to work in television years ago, and I seem to recall that we had to use adapters when fitting stuff that was meant for "standard" (19") racks.

Re:Why invent a new standard?

[Grishnakh]

Because the losses over even 8 feet of rack height can be significant at these current levels. With 48V, you have 1/16 the loss, and you can use smaller bus bars. Copper is getting to be extremely expensive these days, and 48/12 DC-to-DC converters are 97% efficient these days. Or, if this is going to be a whole new standard, why not just run 48V to the motherboards and drives, and convert it to 3.3/5/1.8 or whatever directly? Getting 48V drives may or may not be realistic, but getting motherboards designed for 48V shouldn't be that big a deal for a really huge customer like Facebook or Google. They're going to have to redesign the motherboards anyway for this new physical size.

Re:Why invent a new standard?

[Savantissimo]

The diagrams show the length of the bus bars as being less than 16 inches (40cm). That's quite reasonable, even for hundreds of amps.

Re:Why invent a new standard?

[Shatrat]

21" racks are also a standard size, that is the ETSI standard for telecom racks.

Just about any telecom equipment that fits in a 23" rack also fits in a 21".

Re:Why invent a new standard?

[sjames]

The datacenter as a whole may deal with large currents, but most datacenter personnel are forbidden to touch the parts that use it. The rest of the datacenter tends to run at somewhat safer currents.

That's not to say 208 at 30 Amps won't kill you, but it won't blast molten metal into your face if you accidentally short it.

Re:Why invent a new standard?

[Bill_the_Engineer]

Actually I was thinking of work related injuries involving tools and trying to be humorous.

Re:Why invent a new standard?

[eugene6]

Did that 97% efficiency standard account for removing the waste heat, also? Or was it assumed the waste heat could be freely discharged into the ambient environment and there would be enough "natural air flow" to continue doing so indefinitely?

Re:Why invent a new standard?

[SuricouRaven]

I was going to use 'like a color-coded earthworm orgy' but thought it too disturbing.

Re:Why invent a new standard?

[SuricouRaven]

-12V can indeed go, and should go. 3.3V is used for a lot of important things though. For reasons of energy efficiency, much of the logic runs at 3.3V or less. Including the memory.

Re:Why invent a new standard?

[Grishnakh]

Doesn't matter; this conversion can be done better on the motherboard, just as the converters for the CPU are already on the motherboard (the CPU, since the P4, gets a dedicated 12V feed from the power supply which runs a dedicated converter, probably for ~1.8V). It's more efficient to just put the converters on the motherboard instead of the power supply since the power supply is so far away and the voltage so low, plus then the motherboard can control the voltage directly, lowering it or fine-tuning it if necessary.

Re:Why invent a new standard?

[rev0lt]

I'd consider 400W a bit more than "few hundred watts".
The point I was making is that some applications exist where 12V isn't that crazy. Regarding the story itself, I do agree that 12V DC powering for a rack seems like a dumb idea (voltage drop, cost of copper, catastrophic short-circuits, increase in weight, etc). But I think many modern blade cages (and many industrial computer cages) already provide common DC rails for every card in the system, so maybe isn't such a crazy idea.

Re:Why invent a new standard?

[Anonymous Coward]

Also, if you don't have a lot of space or want to save on soldering you can buy Maxim Line Conversion chips with internal capacitors though they are bit pricy last I checked but if your only making a few protos well worth it IMHO.

Re:Why invent a new standard?

[rev0lt]

And that increased capital cost will be completely and utterly dominated by the increased efficiency in power useage. If you think these guys haven't analyzed the numbers every which way and run a pilot datacenter to prove out their numbers before releasing this stuff you're delusional.

Can you please provide some numbers and web links for that analysis? Because power-wise, you'd save way more switching all the storage of a given rack to SSD than replacing the PSU's with DC rectifiers/inverters. What all these new designs try to accomplish (and not only big companies - every major player in datacenter infrastructure is doing the same) is the same - increase server density by reducing heat emission (and yes, saving power in the HAVAC system). The question isn't DC powering of racks, it's the choice of 12V rails, specially considering there is already a 48V standard for such purpose, at 1/4 of the gauge needed for 12V.

Re:Why invent a new standard?

[Alioth]

The servers still need DC-DC converters, not much runs off 12V inside a server now apart from disc drive motors (and these all have a power supply of sorts anyway, I suspect they have a BLDC motor controller). The CPU (a major power draw) runs off a much lower voltage so you'll still need a reasonably specified power supply inside the server.

Re:Why invent a new standard?

[Alioth]

You don't get to eliminate the power supply inside the server.

Most of what's inside a server runs at 3.3v and below, so it still needs a power supply. Others have pointed out the problems with high currents and therefore high losses at a low voltage like 12V. 48V would have been better.

Re:Why invent a new standard?

[Alioth]

Motherboards don't run on 12V. Virtually nothing on the motherboard (save things like hard drive motors) will run off 12V, virtually all of the stuff on a motherboard runs at 3.3v or lower. The CPU is a major power consumer, and you still need a power supply to provide the other voltages.

Re:Why invent a new standard?

[Alioth]

12V isn't used by much of the server at all. Instead of using 12V then converting to 3.3v, 1.8v, 5v et al. why not use 48v and convert to 3.3v, 1.8v, 5v et al. where most of the actual power is consumed. It'll hugely reduce losses in the bus bar, and make for much smaller bus bars with much lower current.

Re:Why invent a new standard?

[megabeck42]

Your assertion that you'd save "way more" by switching to SSD storage is assuming that the spindle disks are the main consumer of current.

According to WD, the WD20EARX draws 5.3W during read/write, 3.3W during idle, and 0.7W standby/seep(which, admittedly is a rare situation in datacenters.) (from the WD20EARX datasheet)

According to Intel, the Intel 910 series SSD draws up to 25W while active and 8W while idle. The Intel 520 series SSD draws 850mW active and 600mW idle. (from the Intel 520 series product specifications.) I don't know if those numbers for the 910 are a typo, because it seems weird that they'd exceed a mechanical drive.

Either way, my point is that the WD's have a power ratio of 2.65W/TB and the Intel 520 SSDs have a power ratio of 1.78W/TB. Which means that switching to SSDs will save you 33% on your storage power needs. Thing is, because the SSDs have less capacity per SATA port, once you factor in the extra necessary RAID controllers, SATA cards or SATA port expanders, the percent power saving will drop. Admittedly, I have no idea by how much.

I guess, my point is to challenge the popularly regarded idea that mechanical harddrives are extremely power hungry. While CPU efficiency has improved considerably in recent years, I hold that CPUs and associated electronics consume a much larger portion of a server's power than commonly believed.

Also, at idle, the WD consumes 1.65W/TB and the intel consumers 1.45W/TB. Then again, it's not a fair comparison because the SSD can switch between idle and active far more quickly than the mechanical drive. So, once you consider more of aspects of the situation, things become less clearly cut.

Re:Why invent a new standard?

[Vairon]

That's simply not true.

Have a look at your ATX power supply's manual concerning the 20 or 24 pin power supply connector for your motherboard or get a multimeter and test it for yourself. You will see 12V, 5V and 3.3V. Not -48V.

The ATX power supply we all use in our computers is really called ATX12V because it provides 12 Volts to the motherboard. I'm not talking about the 4-pin molex connectors for hard drives. I'm talking about the 20-24 pin power connector to the motherboard. On a 20 pin connector pin 14 provides -12V and pin 10 provides 12V. On a 24 pin connector pins 10 and 11 provide 12V and pin 14 provides -12V.

Here is a wiring diagram that shows this:
http://en.wikipedia.org/wiki/Power_supply_unit_(computer)#Wiring_diagrams

In addition newer motherboards have extra dedicated 12V cables just for the CPU often called a P4 cable. http://www.playtool.com/pages/psuconnectors/connectors.html#atx12v4

Some motherboard require an additional 8pin EPS8 cable to provide even more dedicated 12V rails to the motherboard for the CPU.
http://www.playtool.com/pages/psuconnectors/connectors.html#eps8

Re:Why invent a new standard?

[Grishnakh]

You're right about this: computers these days use a lot of 12V from the main power supply (which converts it from 110/220VAC typically), but then on the motherboard or video card they convert it again to the voltages needed by the *PU. Instead of that, maybe it'd make more sense just to run 48VDC to the rack, and run that straight to each motherboard, where the 48V can be converted directly to the voltages needed, eliminating one conversion step.

Re:Why invent a new standard?

[rev0lt]

Your assertion that you'd save "way more" by switching to SSD storage is assuming that the spindle disks are the main consumer of current.

No, I was only mentioning that power saving by itself isn't the reason for these adjustments. The obvious DC-on-a-rack argument is power savings trough PSU efficiency, but the real problem they are trying to solve is actually heat dissipation and server density, and the obvious AC/DC conversion power savings are only "a consequence".

According to WD, the WD20EARX draws 5.3W during read/write, 3.3W during idle, and 0.7W standby/seep(which, admittedly is a rare situation in datacenters.) (from the WD20EARX datasheet)

I wouldn't use Caviar Green disks on a server, but I wouldn't be surprised if they were common. But, comparing a Caviar Green with a "top of the line" SSD isn't a fair comparison. A Caviar Black/Scorpio Black (still low-end for server storage) will probably consume much more than the model you mention, let alone 15K SAS units. As for capacity, well, if we actually compare server-grade disks, 300GB SAS disks are pretty common. And as a "rule of thumb", using few high-capacity mechanical disks instead of more low-capacity ones is usually dumb, except if you really know what you're doing.

I guess, my point is to challenge the popularly regarded idea that mechanical harddrives are extremely power hungry. While CPU efficiency has improved considerably in recent years, I hold that CPUs and associated electronics consume a much larger portion of a server's power than commonly believed.

A Xeon HP DL160 with 4xSATA7200 1TB draws around 110W working (light load, 1 CPU) and peaks at 160W when booting, due to disk usage and fan usage. We are talking Midline disks (low-end SATA) I'd argue that at least in this model, 40% of the power consumed is disk-related. The spec for my quad-core desktop CPU says it will draw upto 100W of power while in use - that is a (at least) 4x increase in processing capacity when comparing to a P4 with the same power consumption. While server power consumption has been somewhat steady over the past decade, you have a huge increase in processing power and memory for the same watts consumed.

But, my point was that power saving isn't the main goal, given there are so much obvious choices (such as disk replacement) that would result in significant power savings. Heat dissipation is a real problem, and a major challenge that most vendors are trying to solve. As a consequence, usually less heat translates to less power consumed (and less power needed for HAVAC), so it may lead to substantial power savings.

Re:Why invent a new standard?

[Vairon]

You keep saying this but it's not true. Motherboards have several 12V rails because 12V is highly needed by motherboards and the peripherals directly attached to them. For example a dedicated 12V rail is used by a VRM (voltage regulator module), or multiple VRMs, on the motherboard just for converting that 12 volts to the small voltages that a CPU uses. The PCI Express bus also requires 12V (and 3.3V) so the motherboard provides 12V to the PCIe devices. Most CPU fans run as 12V and are supplied via connectors on the motherboard.

http://en.wikipedia.org/wiki/Atx
http://en.wikipedia.org/wiki/Cpu_fan
http://en.wikipedia.org/wiki/Pci_express
http://en.wikipedia.org/wiki/Voltage_regulator_module

Floor tiles

[HockeyPuck]

Nothing like telling the facilities team "ok, we're going with 21" wide racks and you'll need to replace all the floor tiles as well.

Also floor standing equipment (high end disk arrays and floor standing servers) are also made to the 19" standard (either 1-3 tiles wide and 1-2 tiles deep)

I know... this is for greenfield datacenters...

Re:Floor tiles

[Anonymous Coward]

Apparently, the outside widths are the same, so the article says that you don't need to replace floor tiles.

Re:Floor tiles

[atisss]

While on this, let's also replace doors and windows (for defenestration purposes)

Bigger?

[Anonymous Coward]

I'm ok with seeing bigger racks, ifyouknowwhatimean.

Re:Bigger?

[Anonymous Coward]

but 21" makes a rather small rack

wider servers vs 4U box

[Anonymous Coward]

"ready for wider servers?"
Bring em!

Why 21?

[jon3k]

I've got 23" racks in my datacenter right now. We use conversion kits to move the rails in to 19" and gives us extra space on the sides for cabling routing. You lose some space but the quality of life in dealing with the cabling makes up for it in our tiny little space (less than 20 racks).

Why not just change all the bolts to SAE

[gelfling]

that way they demand everyone replace every bolt and make a ton of money for doing even less than re racking everything.

Ok, then.

[Impy the Impiuos Imp]

> Open Compute Developing Wider Rack Standard

Well, programmers have been getting fatter over the years. I suppose this applies to the female ones, too.

Great news!

[demonbug]

I think we can all agree, the bigger the rack the better.

(Obvious joke is Obvious)

Re:Great news!

[Iniamyen]

Especially when they're STACKED!

OK if it's an actual standard

[Urban Garlic]

All will be forgiven if this standard is better than the current scheme. The *only* thing that's consistent about it is the width.

The current scheme has a lot of problems with rail-kits fitting into some racks and not others, because they're too deep, or not deep enough, or because one rack has small threaded holes on the inside of the posts, and another has the big square ones. In my set-up, we only have five racks, and already we're running into problems placing equipment because of differences in the mounting geometry.

Actually, I don't even really need standard mounts, I'd settle for consistent nomenclature -- then at least I could buy adapters, and finally be able to put any piece of equipment in any rack.

12 Volts?

[AB3A]

Why 12 volts? Telephone companies figured out decades ago that 48 volt positive ground systems were more desirable. They reduced the need for heavier copper wire, and they are not likely to be lethal shock hazards (though burns are certainly possible).

Furthermore, every motherboard has multiple switching supplies built in. We have 12 volts, 5 volts, 3.3 volts, 1.8 volts, and probably some adjustable voltages too. Some even have separate regulators for individual parts of the board. We will never be rid of the power supplies. We have simply moved them closer to the processors, memory, I/O, and GPU. Why not design the boards to use -48 volt battery systems as primary inputs so that we can reduce corrosion, use existing infrastructure designs, and keep I^2R losses down?

Oldest standard in computing.

[Animats]

19 inch racks are the oldest standard in computing. The ENIAC used 19 inch racks.

This is really yet another "blade server" scheme. The whole rack is one chassis. Units are on vertically mounted boards with front faces and handles which look like an extra-deep variation on the old Eurocard form factor. That's reasonable enough. It's a lot like 1980s IBM mainframe or 5ESS packaging. Vertically mounted boards are better for airflow, anyway.

It's not about racks which take horizontal boxes like 19 inch rack components, but are slightly wider.

Re:Oldest standard in computing.

[swb]

Maybe that's the whole idea -- develop an "open" blade center scheme that will allow you to mix and match blades without the vendor lock-in that existing blade center systems have.

market opportunity

[Anonymous Coward]

Somebody needs to design and produce kits to retro fit 19" rackmountables in a 21" rack.

RTF spec. - benefits and potential issues.

[gstrickler]

The spec can be downloaded from here

At first, I thought this sounded like a stupid idea too. Then, I read the spec. They're not just changing the width of the equipment area, and it's not just an extra 2".

External width is unchanged from the 19" rack standard, it's still nominally 23.6". No replacement of floor tiles or room redesign necessary.

Equipment width is increased from ~17" (on a 19" rack) to 21", it's allows 4" wider equip.

Power is handled in 3 "zones" per tower. Each power zone provides 12.5V DC power on each of 3 independent pairs of power rails, No AC power supply is required for each piece of equipment, but they will need DC-DC converters and VRs to supply the voltages needed for their specific components. That saves some space on each device, and provides a slight improvement in efficiency. Because this is standardized worldwide, there is no need for each device to have different power circuitry for different countries.

Because a zone can have triple power rails, devices can use 1, 2, or 3 power rails to provide whatever level of redundancy is appropriate.

Space for switches is included in each rack, along with power monitoring/reporting circuitry per rack.

Battery backup power can either be built into the power supply for each zone, or supplied from a separate battery rack.

The specification allows for many AC or DC power sources, this is the only significant part of the spec that will vary by country as the power units will need to support the available AC and/or DC supply.

All devices are to be hot-plug compatible.

So, it does have a lot of advantages.

Here are the concerns I have with it:

All power rails appear to be exposed. While they are on the back, this could be a significant safety (personnel and/or fire) issue. Considering that you can up to 500A @ 12.5V DC running through the zone power rails, and potentially more for the main cabinet DC power rails, exposed seems like a bad idea.

The standard allows depths from 36" to 48". With the way devices connect to power rails on the back, it looks like this means you will need to use devices designed for the specific depth of your rack, or use a shim to extend the device to match the depth of the rack. I believe they should standardize on one, or at most 3 depths, and have a standard set of shims to connect the devices designed for the shorter depths to fit the deeper racks.

Re:RTF spec. - benefits and potential issues.

[megabeck42]

*snip*Here are the concerns I have with it:

All power rails appear to be exposed. While they are on the back, this could be a significant safety (personnel and/or fire) issue. Considering that you can up to 500A @ 12.5V DC running through the zone power rails, and potentially more for the main cabinet DC power rails, exposed seems like a bad idea.

*snip*

That appears to be an illustrative picture. An image from a different article of an "in production" or "active testing" rack shows grounded shields around the bus bars. This is the wired.com article I'm referring to. The picture is somewhere in the bottom third.

12 volt bad. Also, I don't think it's broke...

[Copperhamster]

12 volts is horrid. Really. I have a friend that works in the car audio 'boom boom' industry and he really laments 12 volts. He has to run 0 gauge wire for items that bolt onto a fold down back seat of an SUV... how big a bus bar will we need for a full rack of 12 volt junk?

The limiting safety factor on voltage is breakdown of insulation. The limiting safety factor on current is wire diameter (and to a lesser extent, length). We've converted a good deal of our AC to 240 just to increase wattage capacity. Don't go the wrong way guys.

Also, if you are redefining racks:
23" already exists.
If you must make a new size, make it metric. Also make the rack units metric. And saying 1RU = 2.54 cm is not making it metric.
If you are standardizing everything else, come up with one (not neccessarily new) standard for 'how things are connected to the rack' (one thread, one style hole, one style thing to adapt thread to hole) so I don't have a drawer full of 18 different types of hardware for the things. (ok is this rack 8-32 or 10-32 or 12-24 or M5 or M6. Oh wait the rack kit wants square holes so I need to put this server over in that rack instead... but the nas array wants round unthreaded holes so it has to go across the room)

Re:12 volt bad. Also, I don't think it's broke...

[Anonymous Coward]

If you must make a new size, make it metric. Also make the rack units metric. And saying 1RU = 2.54 cm is not making it metric.

making it metric really only makes sense if you have to constantly convert between different units... who cares if it is 2.54 cm, you are working in units of 1RU anyway.

Re:12 volt bad. Also, I don't think it's broke...

[Savantissimo]

It's a maximum 16" (40cm) run, so no big deal. Really, they could use 8AWG wire (0.1285" 3.264mm dia.)for 200 amps at that distance with less than a 3% voltage drop

12 Volts for a better life!

[beltsbear]

Support for this standard provided by the copper mining and manufacturing association.

METRIC

[bussdriver]

New specs should all be metric; it is not modern if it is using imperial measurement. The new 21" should be 534mm.
Keep the outside stuff the same but convert it all into metric.

Re:METRIC

[gstrickler]

The specs are in metric, with US units for the convenience of those who are metric impaired. You should try reading them before posting.

Re:METRIC

[bussdriver]

No time; but the summary was not in metric either. Some of us do not use that crap system.

Who does't like a "wider rack"?

[rla3rd]

Boobies are fun!

24" Standard

[Anonymous Coward]

Or they could just go to the existing 24" standard.

The numbers all go to twenty-one

[Botia]

Well, it's two wider, isn't it? It's not nineteen. You see, most blokes, you know, will be serving at nineteen. You're on nineteen here, all the way up, all the way up, all the way up, you're on nineteen on your rack. Where can you go from there? Where?

inches to the 100th!

[SpzToid]

Here's an argument for inches I try to impress illustrators I work with; especially in Europe. I show them my valuble-to-me steel 6" precision ruler with each inch etched into 100 divisible units. They are just small enough to actually see and use for measurements that are digitally input to make the illustration.

All the time we need to copy some part-of-something and make a drawing out of it, and that level of precision makes a difference, and eases the work. This trick I learned from older illustrators before me, back in the days before the 2d CAD systems arrived, (with their awful strobe-ing lightpen displays). We used ink on mylar with knives, and we liked it.

I still use that ruler all the time. I don't know where to get another, this baby is gold. It fits perfectly in my plastic pocket pen protector to help me be the fashionable engineer that I am..

Re:inches to the 100th!

[Anonymous Coward]

Any industrial supply house -- mscdirect.com for example -- has those. There's 3 standard graduations I know of that include a 0.01" scale:

• 5R -- 1/32, 1/64, 1/10, 1/100 (seriously, 1/10? waste of a perfectly good edge if you ask me)
• 7R -- 1/16, 1/32, 1/64, 1/100 (really?)
• 16R -- 1/32, 1/64, 1/50, 1/100 (that's more like it!)

There's also some decimal/metric rules with 1/100th and 0.5mm, but they don't have any 2^-n scale -- if anyone knows where I can get one of these, let me know.

Ideally, it would be 64th and 100th on one face, and two 0.5mm scales on the other face (I'd also take 1mm/0.5mm) -- because if the inch scale you want is on the wrong side, you can flip it 180 and read backwards from 6 inches, but the metric scale only lines up to the zero end (6"=102.4mm), so you need it on both sides.

It's not bikeshedding.

[Anonymous Coward]

It's not bikeshedding. 95% of the people on the planet use metric units. It makes sense for the last 5% to start using it too.

Telephone racks are 23", not 21 or 24

[billstewart]

I'm not familiar with a 24" rack standard - 19" racks are usually in 24" cabinets. What uses 24" racks? Telephony equipment uses 23" racks (I forget how wide the cabinets are, but they're wider than 24.) I work in the telecom industry, and we've usually got 23" racks in our labs because when we need a rack there's always some unused 23" rack in the back of a warehouse or storage closet somewhere, whereas getting real 19" rack means dealing with the purchasing department and capital budget. Of course, since routers and computers are all 19" gear, that means we end up buying shelves and stacking stuff randomly on top of them :-) But in practice there's enough non-rack-mountable gear out there that we were likely to do some of that anyway.

My group finally decided to bite the bullet and order a bunch of 19" racks, so we''ll end up ripping out four or five old 23" ones and putting in a couple more cabinets of 19" (and it's not just being cheapass on budget - it also means dealing with the building engineering people to rebuild the earthquake bracing because some of them are different heights, and getting new electricity because we were already pushing our capacity, etc.)

WTF?

[dbIII]

There have been many serious attempts to do engineering projects in metric units

Start with the Eiffel tower and go on from there.
I hate to get personal here, but after reading such silly drivel as the above I really think you've got to get out a bit more and observe what is going on around you before making such a wild and incredibly stupid claim on a site which is actually populated by engineers and scientists.

Re:WTF?

[David Chappell]

There have been many serious attempts to do engineering projects in metric units

Start with the Eiffel tower and go on from there.
I hate to get personal here, but after reading such silly drivel as the above I really think you've got to get out a bit more and observe what is going on around you before making such a wild and incredibly stupid claim on a site which is actually populated by engineers and scientists.

Um, wow. I see you feel strongly about this. Though the tone of your posting makes me think maybe you are trolling, but I will assume that you have simply misunderstood what I was saying.

The Eiffel tower was built in France, so it is not an example of an engineering project in the United States. Remember, I was talking about why the US engineering and building trades have not switched to metric units.

I do not think I am as uninformed as you claim. I am confortable with both measurement systems and fully appreciate the value of base-ten units. I did not claim that the (international) metric system was intrinsically inferior or that the (enormous) cost of finishing the conversion in the US would not be worth paying. I mearly claimed that it is unlikely to happen (in the engineering and building trades) in the forseeable future.

My statements are based both on 1) press reports such as this one: http://web.archive.org/web/20040824125823/http://kennebecjournal.mainetoday.com/news/local/868607.shtml, 2) at least one account I read by an engineer who had participated in US projects in SI units before his firm went back to "standard" units, 3) my person knowledge of US building material dimensions, and 4) my personal experiments in measuring US building materials in meters.

Re:WTF?

[dbIII]

Look kid, while I'm not in the USA I have been a member of ASTM (and also a professional engineer in in my own country) and my strong reaction is due to the sheer uninformed (even about matters in the USA) stupidity of your post above.
As for your point 2; congrats, so you've read one thing by an engineer - read any thread here on any subject and you'll be able to say you've read things by engineers because of some of the people commenting! Silly little "appeals to authority" are useless when you are arguing against someone that is one of those authority figures and doesn't consider that is a strong enough argument.

Not likely

[Immerman]

Assuming the current doesn't pass through your heart, which is particularly vulnerable to electric interference, the damage done will be based on how badly it cooks your flesh, which is in turn based on just how much current is involved. The kicker though is that in most scenarios the current passing through your flesh has absolutely nothing to do with the current the bus bars can provide. The current passing through your flesh will be determined by only two things: the voltage involved, and the resistance of the path it follows. And it turns out that while your body's internal resistance is quite low your skin is actually a pretty good insulator. If I squeeze both probes of my multimeter with one hand, so that the tips are quite close but not touching, I measure about 75kOhms, meaning any 12V source can push only 12V/75kOhm = 0.16mA through my skin, so if I firmly grabbed the bus-bar I wouldn't even notice a tingle, even though it could spot-weld a metal wrench in place. The reason you're warned against doing things like grabbing the terminals of a car battery isn't because it's *likely* to do you any damage, but because it has the potential to do so with a bit of bad luck. If a couple metal burrs poke through your skin into the conductive goodness beneath then POW, barbecue. But then even my ohm meter could potentially kill me if I did something stupid like jamming the probes into my thumbs - the 9V battery can supply more than enough current to stop my heart once a nice conductive path has been formed, heck even a watch battery might do the job.

12v has been limiting car companies for a while

[SuperBanana]

Demands on the vehicle's systems are such that many cars now have their alternators plumbed into the engine cooling system, and are even shaft-driven because of belt slippage. 12V has proven to be a significant burden on carmakers wiring-wise these days with all of the demand for electronics in the car. It wasn't too long ago that cars only had 2 or so computers - the engine control unit, and the ABS controller. I've got a car with almost a dozen - engine, climate, "chassis" (lights, locking, wipers, etc), suspension control, ABS/traction control, airbag, navigation...god knows what else. No DVD player or "entertainment" system beyond the stereo.

Electric steering pumps are increasingly common - they're compact, easy to locate anywhere, easy to control which simplifies hydraulics, and help with fuel economy since there's much less parasitic load.

Even back in the 80's, it was a problem. Audi located the battery in the back seat for crash safety, engine compartment room, and battery longevity (less temperature extremes) - but the cable that goes to the front is pretty damn massive, and failures where they splice a Y off to the main fusebox is fairly common and cause lots of problems because of the voltage drop. The problem was exacerbated by the use of an electric cooling fan which could draw well over a thousand watts - I think the fuseable link by the fan motor is rated for over 100A.

Car companies have been making noises about switching to 24V systems, with a 12V bus only for cigarette lighter accessories...it'd cut down significantly on the size of the wiring needed (which means cutting down on weight!) and transmission losses, both of which will improve gas economy.

Re:12v has been limiting car companies for a while

[rev0lt]

Demands on the vehicle's systems are such that many cars now have their alternators plumbed into the engine cooling system,

Traditionally, (gasoline) cars allways had inverters - traditional (electromechanical) ignition systems usually work at the KV range to generate the ignition spark. They probably also have embedded inversors for any CFL-based TFT panels it may have. Car companies already have the know-how of 24V circuitry, since many manufacturers already manufacture 24V vehicles (RVs, trucks, etc).

But yeah, I agree that 12V for auto isn't the best choice (but 24V will probably only decrease the wiring gauge, not the number of inverters - 36V or 48V would be a better choice), specially with long cables. But that doesn't make it a "crazy" choice - specially considering battery design limitations at the time.

Naturally it is better...

[TheSkepticalOptimist]

because you use the word "open" in front of something and repeat it several dozen times and it just makes it better by Slashdot standards.