Slashdot Mirror
Google Reveals "Secret" Server Designs
Hugh Pickens writes "Most companies buy servers from the likes of Dell, Hewlett-Packard, IBM or Sun Microsystems, but Google, which has hundreds of thousands of servers and considers running them part of its core expertise, designs and builds its own. For the first time, Google revealed the hardware at the core of its Internet might at a conference this week about data center efficiency. Google's big surprise: each server has its own 12-volt battery to supply power if there's a problem with the main source of electricity. 'This is much cheaper than huge centralized UPS,' says Google server designer Ben Jai. 'Therefore no wasted capacity.' Efficiency is a major financial factor. Large UPSs can reach 92 to 95 percent efficiency, meaning that a large amount of power is squandered. The server-mounted batteries do better, Jai said: 'We were able to measure our actual usage to greater than 99.9 percent efficiency.' Google has patents on the built-in battery design, 'but I think we'd be willing to license them to vendors,' says Urs Hoelzle, Google's vice president of operations. Google has an obsessive focus on energy efficiency. 'Early on, there was an emphasis on the dollar per (search) query,' says Hoelzle. 'We were forced to focus. Revenue per query is very low.'"
I think Google may be selling themselves short. Once you start building standardized data centers in shipping containers with singular hookups between the container and the outside world, you've stopped building individual rack-mounted machines. Instead, you've begun building a much larger machine with thousands of networked components. In effect, Google is building the mainframes of the 21st century. No longer are we talking about dozens of mainboards hooked up via multi-gigabit backplanes. We're talking about complete computing elements wired up via a self-contained, high speed network with a combined computing power that far exceeds anything currently identified as a mainframe.
The industry needs to stop thinking of these systems as portable data centers, and start recognizing them for what they are: Incredibly advanced machines with massive, distributed computing power. And since high-end computing has been headed toward multiprocessing for some time now, the market is ripe for these sorts of solutions. It's not a "cloud". It's the new mainframe.
Javascript + Nintendo DSi = DSiCade
From 2007, the modular data center patent (where the bottommost image of the article comes from). There's no lack of patents revealing piece by piece how their power management setup works.
Ah, the catch--22 of the patent--being forced to reveal your hand in order to protect it while underpaid workers at Baidu figure out how to integrate your ideas into their hardware.
My work here is dung.
Get that man a beer.
Get some mainframes.
For cryinh out load, with 1 mainfram you can't have a mainframe with 30,000 or more intances of your operating enviroment on it. Possible up to 100K.
Put 5 of these in each data center. Cheaper to power, you would only need a few people to keep it running, it would run for 20 years.
Want to save more money, here is another way:
Build your data center in the desert and build 150 MW industrial solar thermal system to power it. Sell the extra power.
Oh, and if you are not up to date on Big Iron, don't fucking reply becasue your going to look like a fool.
The Kruger Dunning explains most post on
I'm no guru of servers, but from my own limited experience in installing servers at the small to midsized company I work at, space is always a looming issue. And shrinking the size of the UPS you need can only save money and space in the long run; which any IT manager will tell you is a huge benefit and a great selling point.
Nothing to do but wait for a finished product at this point though.
Posting with out proof reading since 2001.
Greater than 99.9% efficiency? They likely made a mistake in their measurements.
Excuse me, but please get off my Pennisetum Clandestinum, eh!
We all know the searches are actually being done by a large amount of people in suspended animation, being fed the corpses of the previous people.
The thing about each server having its own battery is a cruel joke.
The in-computer onboard UPS is not a new idea. I don't see how they could have gotten any patents on it since I used it have one of these (my day might still). The device I saw had a gel cell mounted on an 8-bit ISA card, full length. It had +5/12v pass through connectors for powering the drives and it powered the computer through the main bus. There was more logic to it, as it had some monitoring capabilities too.
What's next, patenting a hard drive on a plugin board? Been there, it was called the Hard Card and put a 20mb HDD in an 8 bit full length ISA slot, a truly neat idea for upgrading old XT computers back in the day. You could make them work with AT computers too by putting a regular disk controller, without a drive connected, on the bus too and the BIOS would see the XT controller and boot from it.
When the weather gets warmer, Google notices is that it's harder to keep servers cool.
Brilliant journalistic work there.
Google is basically re-implementing the efficiency that already exists in a laptop. In fact, laptops outperform rackmount specifications if not for the fold-open HID and effeminate socket to an external resilient power-brick. What is more advanced than Google's software on Google's hardware is a stack of laptops running DNS for a Freenet backend, stationed at the peak of a mountain like PYRAMID LAKE California with a bat of "wiffi" 802.11abgn serving minimal free internet access throttled down to 2Kbps.
You want to know what's more advanced than Google? My middle fignerrrr.
Googles secret is that all there computers have battery.
I think, it is called a laptop.
...why desktops didn't have a built in battery deal that lived in an expansion bay. If you could even keep RAM alive for extended periods even with the machine shut down that would be spiffy as an option, let alone as a little general UPS.
look at the date the article was published.
"It would be wrong to refuse to face the fact that everything is fundamentally sick and sad."
This is a questionable number. The best DC-DC conversion is around 95% so they aren't including voltage conversions from the battery to what the system is actually using.
wouldn't trust them any further than I can throw them
Given the reliability, it's likely that someone has already measured that particular parameter for you. Have you checked the data sheets?
I am one of many. My idea is not unique, nor do I expect my voice alone to sway you. I speak in a chorus of opinion.
Many data centres expressly forbid UPSes or batteries bigger than a CMOS battery in installed systems - because when the fire department hits the Big Red Button, the power is meant to go OFF. IMMEDIATELY.
So while this is a nice idea, applying it outside Google may produce interesting negotiation problems ...
http://rocknerd.co.uk
I'm working on a solution. If only I can contact Oracle.
You're probably thinking "man, these things are just too big, no one will want one for their home" but in a few years these things will be on everyone's desktop. Sure, the first few desks will be crushed, but I'm 100% sure they will make them fit nicely into your cubicle.
From the diagram it looks like they just need to put a chair in there and you're good to go. Now to compile Counter Strike for this thing.
Anyone concerned that when a SLA batter is charged, hydrogen is one of the by-products?
April Fool?
wouldn't trust them any further than I can throw them
Given the reliability, it's likely that someone has already measured that particular parameter for you. Have you checked the data sheets?
No, I haven't. But I choose, in this case, to predict the future based on past events. It's not an investment so I feel safe doing so. Many computer users do the same - if you've had a bad nVidia graphics card or a bad ASUS motherboard people tend to shy away from buying those particular brands again. With storage being as cheap as it is combined with the importance of reliability I'd rather stay with a brand like Western Digital and, up until recently, Seagate than use a brand that has had performance and reliability issues in the past.
"This food is problematic."
Hundreds of thousands of servers == thousands of dead batteries each month, since those batteries don't last more than a few years.
Now I'd think their design could be gentle on the 12V batteries, since it's possible to design UPSes that don't murder batteries at the rate cheap store-bought UPSes do. But still, they must have an army of droids swapping out batteries on a continuous basis.
Or maybe they are more selective, and only swap out batteries on hosts that have suffered one or two outages. It only takes one or two instances of draining a gel cell to exhaustion before it is unusable.
Give a man a fish and you have fed him for today. Teach a man to fish, and he'll say "WHERE'S MY FISH, YOU IDIOT?"
I am intrigued at the idea of a battery as the power supply. This means you can use a smaller inverter with high quality components and it will produce much less heat. With the supply being a battery, I would imagine you have much less worry about ripple as well. I wonder what happens if a battery explodes?
I would guess they are measuring battery temperature as well as other temps. Batteries only explode when they get really hot. So I would imagine the machine turns off before it heats up too much.
They're using their grammar skills there.
This is composed purely of commodity parts. The power supply is the same thing you'd buy for your desktop, those are SATA disks (not SAS), and that looks like a desktop motherboard (see the profile view where all the ports on the "back" are lined up in the same manner they would need for a standard desktop enclosure).
Only the battery is custom (or even non-consumer grade), and you can note that since the power goes through the PSU first, that's DC power. DC is significantly better than AC, since the PSU then has to convert AC-to-DC (which wastes power and generates needless heat). While you can get DC battery supplies for server-grade systems, these are not server-grade systems. Built-in DC battery backup therefore affords them the ability to keep the motherboards cheaper. Very smart.
Also, if you recall from a few months ago, Google has applied pressure on its suppliers (I'm not sure why Dell comes to mind...) to develop servers that can tolerate a significantly higher operating temperature (IIRC, they wanted at 20 degree (Fahrenheit?) boost). I wouldn't be surprised if the higher temperature cuts down on operating expenses more than smarter battery placement.
Use my userscript to add story images to Slashdot. There's no going back.
Dawned on me the other day how little innovation occurs in our industry EXCEPT by hungry companies. For example, Desktops and Laptops have not really changed, while both have a piss poor design. ABout 4 years ago, it dawned on me that a much better way to design these is to merge them. Basically, different cases where the laptop has keyboard and a monitor hookup while the desktop is sans the prior. The smart move is to move the battery OUT of the case and into the power supply. Right now, you do not get to buy variable amounts of batteries. But a company would do well to sell an external power supply with varying storage capacities, but with a simple 12V line. In this fashion, ppl can pick the parts for a laptop similar to a desktop, while the desktop gets to take advantage of the drop in prices of the laptop linage.
I prefer the "u" in honour as it seems to be missing these days.
Peter Huber in his book on energy policy introduces the concepts of the "energy pyramid" and "energy refining". The thesis that new forms of energy technology use more technology and are subsequently more useful. The pyramid levels include wood, coal, petroleum, electricity, computing and optical. When I read the book a few years ago I always found it curious that he included computing in the pyramid. But I hear about aggregate gigawatts of hundreds of mass server farms in the world, it may start making sense. The web has transformed human technology and the server farms are the battery of the web. When Huber wrote the book he used the example of the automobile as it started being mostly petroleum energy, then acquired more electricity sub systems, and now more computing.
"I worked 14-hour days for two and a half years." -Ben Jai, Google engineer
And I thought these guys were supposed to be smart?
This is an April fool's joke. Look at the date of the article.
Also the two hard drives are plugged into each other.
Still an interesting idea though.
He used a car as a metaphor to explain technology? Brilliant!
What's his /. userid?
look at the date the article was published.
I can't believe how far I had to scroll down the comments before someone realizing that.
"Google's designs supply only 12-volt power, with the necessary conversions taking place on the motherboard"
This seems to be a more interesting point than the battery part. 12V-only?
This means that there's some serious power conversion done on each of the motherboards, and with SMPS evolving at the rate that it is, this could be relevant to anything larger than a laptop.
How much exactly is gained by making such a big change, to a point where you'd need to redesign all of your motherboards, each time for each different chipset? (they mention they use both Intel and AMD)
Will this particular change make it into desktops? How much *more* efficient would it make the overall system?
Entomologically speaking, the spider is not a bug, it's a feature.
Why is it that Google doesn't have to worry about FCC, CE, or UL safety and EMC regulations? And why are they using a RoHS prohibited battery which uses lead???
I'm a little surprised by the keyboard and mouse port and the two USB ports. If it uses USB, why not just use that for the keyboard and mouse? And why the second USB port? I suspect the second port doesn't consume extra energy directly, but it causes air resistance where they'd like to clear path to drag air across the RAM and CPUs.
And why the slots which will never get used? In quantities like Google buys, you'd think those would be left off.
Maybe they don't make any demands on Gigabyte (the manufacturer) and just buy a commodity board? When they're buying this many, you'd think Gigabyte would be happy to make a simpler board for them. On a trivial search, I don't see the ga-9ivdp for sale anywhere, but maybe it's just old.
So this sounds like one of those "so obvious, no one thought of it" questions - if Google is so concerned about precious mW that it standardizes on 12V hardware to reduce current losses of sending 5V & 3V power from the powersupply to the board, why do the CPU's have fans???? The side view of the chasis seems to suggest that with a few minor tweaks the units could rely on passive cooling and use the data centre / container fans for air flow.
1) Move hotter components like the CPUs to the front and replace fans with larger passive heat sinks.
2) RAM modules lined up to ensure proper airflow to the back of the chasis, chipset heat sinks lined up accordingly.
3) HD's laid over top of voltage regulators with appropriate heatsinks
4) power supply and battery at the rear.
Have the hot air return duct work arranged at the back of the rack with appropriate holes and seals so that the units make a good connection to maximize airflow.
I'm in my right mind and I have the answer to everything!
Google is basically re-implementing the efficiency that already exists in a laptop.
You have a laptop with >1000 processors, consisting of several times that many cores, with its own built-in gigabit ethernet running on built-in gigabit switches?
I'd hate to sit next to you on an airplane!
Javascript + Nintendo DSi = DSiCade
I believe the joke was that the distance a DeskStar can be thrown may be published in the data sheets. Being such a common concern and all. :-)
Javascript + Nintendo DSi = DSiCade
You don't need to meet regulation safety if you are not selling them, just for your own use, and you worked it out with your insurance company.
It's ok, appearently he stores it in his middle finger.
Date Center Knowledge has videos of the secret server and a tour of one of the container data centers.
Whoosh!
Clearly Google's entire business model has failed because of your insight!
Read the Google paper on hard drive failure. They may have thought about things.
...imagining a Beowulf cluster of these?
Aww - nevermind.
Once upon a time, maybe 6 or 8 years ago now, I got to sit down with the CEO of APC and basically told him I wanted battery backed in-computer power-supplies, something small yet efficient. I wanted functionality like my laptop does, unplug PC, move it, plug it back in. Same for my servers (might have been when that whole Netshelter product line started up.
:-)
Ah, too bad I kept no notes, no logs, could have made a fortune suing Google.
Probably because they own these datacenters and can do what they want with them. The EM emissions are probably contained by the fact that the servers are all in a giant metal box. UL is optional, and if they don't want to go through it they don't have to. It's not like they're selling these servers to anyone.
I read the internet for the articles.
My servers also have 12V batteries for a long time now. I could not imagine someone could have a patent on this, though, it's a very simple mod everyone can do. I also run my PDA and netbooks using AA batteries. Can I get a patent for that?
There's only so many places you can connect a battery to a PC and all of them have already been implemented by someone at some point. There's been motherboards with second power connectors, motherboards with battery connectors, power supplies with batteries, power supplies with battery connectors, DC power supplies connected to external batteries, integrated UPS systems which take in and put out AC and which are basically just hooked up in line with the power supply... Off the top of my head I immediately think of AS/400 systems which were offered with integrated UPS before they even renamed it to zSeries or whatever it is. (I always forget. AS/400 was a good, IBM-sounding name.) The solution which comes immediately to my mind for a google-style distributed data center would be to use something power-efficient hooked up to a PicoPSU hooked up to a SLA battery hooked up to a charger hooked up to your power source. Cheap, simple, and built with commodity parts. (They seem to sell a UPS charger unit where you can get the PicoPSU as well.)
"You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
I hope the batteries are not made by SONY.
Because they are not selling these products.
Basic FCC for business would be the only actual government requirement. All of the other's are optional 3rd party certifications and not required to make or import products into the US. Also RoHs has exceptions for batteries.
I've worked on UPS projects for large tech companies and their standards were always higher than the basic certifications. There is no point to waste the money on 3rd party certification for in-house use products.
I could design this PSU configuration, and I do electronics only as a hobby.
First, your main PSU delivers 12V in this scheme. Then this is stepped down to 5V and 3.3V for mainboard use, a design that is already employed by some Enermax PSUs, for example. For the 12V line, remember that +/-10% lower is acceptable. The lead-acid battery delivers up to 14V, so you need a step-down converter to 12V. In fact, you can design a switching regulator that steps the input voltage down to 13.2V (12V+10%), if it is larger, and just passes it through for 13.2V...10.8V with very, very low losses. A similar design can be done for 5% tolerances. Modern switching FETs go down to 4mR per transistor and you can do the transition from switching mode to pass-through mode very easily, e.g. with a small microcontroller that can then also do numerous monitoring and safety things. I had actually considerd such a design (purely analog though) for a lower-power, 12V external supply system myself some years ago, but a single UPS was so cheap that I did not went through with it.
I do not mean to belittle the what the Google folks do, though. The real ingeniuity is relaizing you can do it this way on a datacenter scale when nobody else does it. The engineering is then not too demanding, at least for folks that know what they are doing.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
Arguably, APC has become a mainframe vendor. They sell rack systems with integrated power, cooling, and cable management. Add commodity motherboards, CPU parts, disk drives, and software, and you have a mainframe. It's not that different from what HP or SGI or IBM or Sun will sell you. Especially since the "mainframe" vendors have mostly moved to commodity CPU parts.
I've pointed out before that computing is becoming more like stationary engineering. Stationary engineers run and maintain all the equipment in building basements and penthouses. With containerized data centers, computing looks more and more like that.
How does it hold up against an EMP?
I think a lot of people are missing the obvious trends.
1) people are moving towards mobile phones and netbooks for mobility's sake
2) people are weary of crappy battery life
3) mobile data connections are becoming ubiquitous
This all combines to form the perfect mix for long battery life dumb terminals that only have a screen, camera, network connection and maybe a graphics processor/keyboard/voice. What's the best way to boost device-battery life? Remove computing from it.
So this cloud computing thing, whether its a mainframe, cluster or just plain SETI@home style network, will take over the functions normally performed by your laptop/mobile/desktop. Think about it... its already happening (nokia's internet tablet, etc.) and major corps are preparing for this move.
Read what I mean, not what I wrote.
With UPS:
1. powered by a bunch of 12 Volt batteries
2. circuit pumps 12 Volt DC to 120 Volt AC 60 Hz
3. cables carry the 120 Volt AC 60 Hz to a computer
4. switching power supply unit in the computer converts 120 Volt AC 60 Hz to DC, followed by high-frequency AC, pushed through a transformer again to bring down to voltage.
5. Output of the transformer is regulated to 12 Volts, 5 Volts, -5 Volts, -12 Volts, 3.3 Volts etc
This is not something new... you remove steps 2,3 and 4 from what everyone is already doing and suddenly you have google's method.
I'm almost certain that's for cost reasons. Sure, Google could probably get Gigabyte to custom-make a board - but then they'd have to pay that much extra to custom-design it, and Gigabyte would probably charge them a little bit more. As it is, they can just use the same lines that Gigabyte is already running, and get the same hefty discount that Joe the Computer builder gets from the massive volume they're running.
The power of a thousand cores is no match for my asbestos underpants.
Relax. Before you know it, all those deathstars will be replaced with SSDs and each server will have a gerbil running in a cage to power it. The gerbil bio waste will be routed to the ground and that crane will pick up each shipping container to rotate it to a fresh spot of grass every week, while the freshly fertilized patch is used to grow organic vegetables for the googleplex.
I thought the onboard voltage converters were Google-specific? In that case, they must have a reason for the PS/2 sockets. Mind you, it might just have been more expensive to redo the board design to exclude them than they would save in parts.
Reality is the ultimate Rorschach.
Google's designs supply only 12-volt power, with the necessary conversions taking place on the motherboard. Googles data center efficiency has been improving gradually. That adds $1 or $2 to the cost of the motherboard, but it's worth it not just because the power supply is cheaper, but because the power supply can be run closer to its peak capacity, which means it runs much more efficiently.
From that I assumed it is a custom board since I'm not aware of any boards out there that doesn't rely on 5V to be delivered by the PSU.
If you look closely at the top view, there are many components missing. most noticeably, the graphics connector. They used standard boards, but left out the unneeded components, which would be easy to for large orders. USB and network probably use the same chip, so leaving out USB doesn't save them much. I don't think the connectors are very expensive in bulk, or this is en early prototype.
From the article, the board seems to be non standard and accepts pure 12V dc as power, converting for peripherals. Whereas ATX PSUs provide 5v rails, and maybe another voltage, I forget.
As for the ports.. who knows what sort of monster KVM setup they have for all those nodes in a container. Perhaps PS/2 provided them with less complexity and more robustness compared to USB.
Seriously? Why all the hardware when you could virtualize? Or are they really using all of the processing power so efficiently as to not need virtualization?
Wait, my laptop has one of those too...
In other news, is anyone else surprised that a built-in UPS is so slow to catch on for the desktop when notebooks have had it by definition for years? Sure, powerful batteries are expensive, but you'll wish you had one when a power blackout destroys half a day's work. It's one reason why I hesitate to get a desktop PC.
This is another story that proves that small and medium size companies should not buy their own server hardware anymore. The economy of scale found in Google and Amazon sized companies can drive the hardware innovation like on-board batteries. Everyone else should just lease time on those systems while the big boys race ahead on hardware innovation. Buying your own server these days is an exercise in planned and accelerated obsolescence.
It was actually leaked years ago about the battery. It's just the first time Google is talking about it. For example: http://www.webhostingtalk.com/showthread.php?t=492779 This is like Israel's nuclear capabilities. Everyone knows they have it, but they officially decline to discuss it.
Well, I'd hate to be ON that plane.... that system would be a cloud *in* a cloud... until the plane crashed from all the weight it.
Funny... captcha is "kerosene" (which some planes use, IIRC...)
Previously: "Linux... Toward the Sunrise..." Now: "Linux... Toward the-- No, now, part of Every Sunrise"
Removing the PS/2 ports and the PCI and PCIe (or is that AGP? sorry) slots would not require a different board design, just a different assembly option (i.e., you tell the factory robot to skip soldering in those components). I don't think that should cost extra when ordering large quantities of boards, and the parts Gigabyte doesn't need to manufacture/acquire and assemble will save everybody money. So I'm with the GP on this one.
I guess he was talking about the battery which provides built-in UPS to a laptop.
RTFA, it is not about processors but the innovative UPS-in-server model.
Actually, they're not.
Laptops run slower than their PC counterparts.
Laptop drives run slower than their PC counterparts.
Laptops run hotter under load than their PC counterparts.
If you look carefully at the picture, they've found a 12v motherboard, tied a 12v battery directly into it, and used otherwise commodity parts. That's been the mantra for Google for as long as I can remember. Oddly enough, that was my mantra when I built up a big network. Lots and lots (and lots and lots) of cheap servers are better than a handful of really expensive ones. That saved our cumulative posteriors on more than one occasion.
I've spoken with some people who have personal knowledge of Google's equipment. They were setting up with RAID 01 or 10. I suspect with the two drive configuration, they're only setting up with RAID 0 now, and the redundancy is across multiple servers. I can confirm that they are using this open tray system for it's superior cooling.
I had considered open trays like this, except there's one huge downfall. You would have to be amazingly careful of what happens near the rack. If you are screwing something in, and the screw or screwdriver falls, that can become very bad very quickly. Did you see any fuses or breakers from the battery to the power supply?
Short of making the area around the rack a metal-free zone (no screws, screwdrivers, rings, keys, watches, etc), you'd seriously run the risk of shorting something out. I know I've been working up in the higher areas of a rack, and dropped screws. You listen to it rattle it's way down across several machines until it finally hits the floor. Since I use closed servers cases, it's never a problem. Maybe they don't have a big problem with it at Google, but I'd be terrified of it. Anyone who says they've spent any substantial time working in and around racks, and haven't ever dropped anything, are lying. I do love the idea for free airflow and better cooling, but ... well ... I like to keep magic smoke in it's place. :)
The one-battery-per-server is a nice idea though. I may look into that for future builds. Most PC's have 5v and 12v output. That power supply only indicated a 12v output, and didn't have any wires that indicated anything different.
Serious? Seriousness is well above my pay grade.
All notebooks have built-in battery right?
I can tell you why. The priesthood of the mainframe, I call it.
I started working as a junior programmer at one of the "Baby Bells" in 1990. At the time, the CICS/MVS environment contained most of the data. However, whenever someone asked for any new change, the answer came back: 3 months, $100,000. So the execs starting hiring pc database guys like me. We used 386s to run clipper/dbase apps, downloading reports by capturing the ascii using telix or telemate, then using the clipper/dbase code to cut up the ascii and process it in batch jobs that would take overnight on the 386. All because the mainframe guys had an attitude. O remember once a sales director asked for an "average sales per day" column in a monthly report, literally dividing the sales column by 20. The answer: 3 months $100,000." I had it for him in 2 business days, and that was because I was busy planning the department's Christmas party at the time.
The PS only provides 12v, according to the article the motherboard itself does the stepdown to 5v where necessary (i.e. custom motherboard built to spec). Though I do hope this becomes a standard.
And I'm guessing he didn't RTF Post he was replying to.
I always thought it was funny that they required several voltages. It would seem like it would be more ideal to make everything work to one specification (like 12VDC). I know the voltage regulators drop the voltages as needed for the CPU, but we've been using those for many years now. Not a big deal, but a waste of power. In a single PC, it's not a big deal, but when you're working in groups of 1000 machines, it becomes a bigger problem.
Serious? Seriousness is well above my pay grade.
...and every day, the world gets one step closer to being able to do just that... reference
There are voltage regulators that can drop/boost the voltage to a predetermined voltage and do so with 90+% efficiency. Look for 'buckboost regulator' or 'switching regulator'.
Tom.
The problem I have with running a motherboard directly from a 12V battery is that most batteries are 12V nominal; actual voltage varies quite a bit (10.5-13.8 for a Pb-Acid). So the question is how well do the 12V components cope with the lower/high voltage? Most of the logic should be OK; that's all 5/3.3/1.xV. I'm guessing the only stuff that really uses 12V anymore is actually disk drive circuitry(not technically on the MB).
I have a suspicion that you really don't want to be running a hard drive off a voltage supply that varies by up to 25%. They must have solved this somehow (step up + step down converter? But that is not efficient) but I really see no point in using 12V motherboards unless everything else can reliably run off the battery first. The home consumer may as well stick with getting 5V from the PSU and letting that dissipate the heat from the step down conversion until we're all using 5V disk drives. In which case, we can probably move to lower voltages (and lower voltage batteries); ~8V seems about right to get a stable 5V.
I have determined that my sig is indeterminate.
That's still only 90% efficiency. If they didn't have to change the voltage to operate, that would be out of the equation entirely. :)
Serious? Seriousness is well above my pay grade.
Because its their datacenters, and they can do what they want. Remember private property and private enterprise?
I want to delete my account but Slashdot doesn't allow it.
http://news.thomasnet.com/fullstory/809646
Probably getting to be standard for small form factor mobos; space, power and noise are definitely on people's minds.
Give a man a fish and you have fed him for today. Teach a man to fish, and he'll say "WHERE'S MY FISH, YOU IDIOT?"
Modern high speed chips (which draw the bulk of the power in a typical PC) run thier core logic at much lower voltages. Typically somewhere between 1V and 2V though I think some may have gone below a volt now. Theese very low voltages have to be produced very close to the chip that uses them to avoid huge losses.
This means that modern PC motherboards take most of thier power at 12V anyway. The 5V and 3.3V lines really only serve to power the low speed chips and some of the interfaces between chips.
Given that I doubt there would be too much efficiancy loss from making a 12V only board. You could probablly even design it to hapilly deal with an input that was only approximately 12V without losing too much (since most of that 12V power is going to the input of switchers anyway).
note: i'm known as plugwash most places but i screwd up registering that here somehow in the past and now can't register
Looking closely at the top and side pictures, I can identify PS/2 Keyboard/Mouse connectors, a serial port, and a metal can that seems to hold 2 USB ports and maybe a firewire port (the top connector-Not an RJ-45).
Next to the can is where a dual Ethernet (standard on most server MBs) connector would go. Nothing there.
Are they using USB/firewire for communications or add-in boards (maybe Ethernet or Infiniband)? Or maybe it's a Zen thing...
There was a company that used to advertise in the old phone book thick computer shopper mag, that had a 12V battery and UPS right inside the AT power supply (The old large odd shaped power supplies for 'tower' cases, that were otherwise mostly empty)
I'm guessing that had to be no later then 1993 that I remember seeing those.
Anyone notice in the video, the 8 or so boards of what I assume is memory? I'm not a hardware person, but they didn't look like the normal DIMM boards, so maybe they are something else.
And I didn't see any disks (but I only watched once). Maybe they run entirely on ramdisk, which would explain the importance of the battery.
Do as you would be done to.
LOL. DamnStupidElf, you are a very smart elf. A large percentage of people who work at technically-oriented companies use corporate-speak, indicating that they don't really understand what they are selling, and don't care.
RoHS does not cover batteries. Lead batteries are covered by recycling laws in most places and I'm sure Google follows those.
Why redesign the boards, instead of just not soldering those parts on? Saves the labour and parts. I don't think there is any logic in those parts that the board would miss suddenly.
If you look though, I don't think there are many screws in that design, both the power supply and the hard drives seem to use a velcro fixed strap, which looks to be an inspired idea. And I've heard before that google don't service servers in place, if one goes bad, they simply pull it out, fit a replacement, and send the bad one off to a dedicated repair desk.
I remember talking to a VA Linux sales guy in the early 2000's who was kind of ticked because he tried to sell Google on VA's kick arse (for the time) servers but google wouldn't have any of it. According to the sales rep Google was using racks filled with 1U sliding keyboard trays and just laying a motherboard, power supply and hdd in each try.
Google's idea was that if you need thousands of identical servers why bother with a case and all the fanciness of a commercial Dell/VA/HP type system. If a component dies just swap in another shelf.
Seems like the same idea only in a shipping container. The shipping container this was also mentioned on /. a few times so I guess the the FA does read /. :)
I was at DEC when we brought out the first VAX and some areas were enforced "metal free". You shouldn't wear a metal wristwatch strap when working near a 5V power bus that gives 300 AMPS !
I wasn't quite as concerned with the screws of the machine, as I would be of say a rack screw. They may use pop-in or fixed tray. Since they have a fixed configuration, I can't imagine they move the rack configurations around much.
But, what about patch panels, or any metal accessories at the top of the rack. As far as that goes, someone may have a note in their pocket saying what to work on, and they pull it out, along with some loose change, which could go bouncing. See, unless it's a truly metal free environment, bad things could happen.
Hell, as far as that goes, someone like me who wears glasses, and doesn't have thick plastic frame glasses, could bump into something, the glasses fall off, and fall into a sea of open cases.
I do really like their velcro method though. It should provide the electrical grounding required of the components (they still touch), while ease of repair. I would suspect when they have a failure, it's failed over to another standby unit, they yank the bad unit and bring it to a workbench, and throw in a good unit. The workbench time would be cut down substantially.
Their design actually reminds me of my "test machine" from when I worked in a computer store. We had parts laid out on a wood board, where we could swap any suspect part into it, and run it in a real machine. Ours was spread out a little more, but it was a different environment. :)
Our test board was left powered off when we weren't using it, and safely put above everything else (up on a shelf). When we tested with it, we used it in a clean area of the shop, where people weren't passing by with stuff. (like screwdrivers, loose change, etc)
Serious? Seriousness is well above my pay grade.
As Tesla proved, it's not the volts that kill you, it's the amps. :)
Ya, there are many caveats to that. (>50 volts to overcome the resistance of dry skin, tesla used very high frequencies, etc)
I've touched both terminals on a group 8D battery (12V 1200A) just to show that it's safe. It usually freaks kids out, especially when I've just spent several minutes explaining that electricity kills, etc, etc, etc.
A lot of it is the load though. In high school, a physics teacher told us the story of a kid who was jump starting his car. It was told as the student went to the school years ago. Ahh, the makings of an urban legend.
Anyways, he was a football player, and his class ring accidentally touched between the terminal and the final connection. He said the ring immediately welded to the terminal and the cable, and due to the resistance of the ring, it became very hot, and melted the skin where he wore the ring.
It may have been true. It may not have. But, it was a valid learning experience. It's the same reason that you're warned not to put wrenches near the battery. It can (and will) arc weld itself in place, and be difficult to remove.
I've never done either the class ring, nor the wrench, but I've touched wires to the terminals (on purpose, mind you), and they've stuck because the arc was hot enough to melt both ends, making a bond.
I have a very healthy respect for electricity. We have a long and sorted history together. Just talking about it makes my hands tingle (a few oops's over the years), none fatal, obviously.
So ya, shorting 5V 300A could be a very bad thing. At least enough to let magic smoke out, and upset some very expensive parts. :)
Serious? Seriousness is well above my pay grade.
I'm glad you have a healthy respect for electricity.
Continue your relationship.
When your hands tingle, she's just warming up.
Obviously.
We've been dancing very nicely together for many years now. I've done some higher voltage playing, while paying extra special attention to safety. :)
I've made a few things heat up, but they've been kind enough to be melted wires/boards and popped circuit breakers. :)
Serious? Seriousness is well above my pay grade.
Nylon screws. Plastic screwdrivers. NO BEVERAGES.
But can you mandate no wedding rings, no hair clips (usually women only), no pocket change, no pens with metal clips. No ... well, run them through a high sensitivity metal detector before they go into the DC?
I've actually been bad about the drinks. One facility didn't care, since I only drank in my own cage. It was my problem, not theirs. We had plenty of screws that would inadvertantly bounce off of the top of live servers, and rattle their way down to the floor, but my worst drink incident got my table (in my cage) wet, and almost (almost) got my laptop wet. The biggest problem was, I didn't have anything else to drink until after I left. You'd have to be just plain dumb to put an open drink container on top of a server.
Serious? Seriousness is well above my pay grade.
You are completely blowing smoke. A top-end modern mainframe does NOT contain ANY of the following:
1) Commodity Motherboards. In fact, they don't contain anything that even resembles a motherboard. (It would be impossible to make a fault-tolerant machine if it did.)
2) Commodity CPU parts. Okay, they MAY contain a CPU chip that exists in smaller units, but that is about it. However, many mainframes use completely custom processors. Not even the heatsink the CPU attaches to is anything you would recognize.
3) Normal Disk Drives. Very few, if any, top-end mainframes contain more than a small number of disk drives, if they contain any at all. (Many do not.)
4) Commodity software. You have to be joking. No mainfrmae will boot Linux natively. Even if it could, it would be a waste of some extremely expensive hardware.
5) Standard power supplies. Top-end machines use 3-phase industrial power connected to a wire thick enough to run several houses.
The whole setup is then built to be so fault tolerant, you could literally run a bullet through any single component and it would not even hiccup.
If you ever get the chance to go into a large bank's or large government agency's data center, ask if you can peek inside the front and rear covers of the zSeries mainframe they almost all use.
It looks absolutely nothing like a rack of servers. Nothing at all.
SirWired