Google Patent Reveals New Data Center Innovations 82
miller60 writes "'Google is seeking to patent a system that provides precision cooling inside racks of servers, automatically adjusting to temperature changes while reducing the energy required to run chillers.' The cooling design uses an adjustable piping system featuring 'air wands' that provide small amounts of cold air to components within a server tray. The cooling design, which could help Google reduce the power bill for its servers, reinforces Google's focus on data center innovation as a competitive advantage. Check out the patent application and a diagram of the system."
Re: (Score:2)
Right, let me do all the work while you reap the benefits and I continue to eat moldy bread.
Re: (Score:1)
There's no reason both models can't co-exist, but neither is a replacement for the other
Re: (Score:1)
Bollocks.
Data center innovation should benefit whoever innovates. If Google are 5 years ahead of any other DC operator in terms of cost savings, that should be incentive enough. Preventing any other company from also saving power by patenting things like ducts+servomotors is not only counter to the interests of society in general, but counter to the market's health.
First mover advantage should be enough. If you need more than that then you're not really innovating. Patents just allow companies to innovate *
Re: (Score:2)
Google does release many of these inventions to the public for free.
For instance, Google released their designs for their own special power supplies. Apparently they run DC data-centers and reap massive energy savings.
Re: (Score:1)
Google only releases those designs that don't allow a competitor to interfere with its business. Before any result of its R&D is released, there's a decision making process that occurs, where they ensure that the item in question does not confer any significant increase in capability to its competitors. Typically, they just open source those things that their competitors already have proprietary versions of (such as MapReduce and BigTable) so that they don't bring about any change in the competitive lan
Re:No matter how innovative (Score:4, Insightful)
Google is a for-profit company that does a lot of work for open standards, FOSS, and charities that they don't have to.
You insist they must do it for PR, yet their PR campaign has been so successful with you. They don't even go out of their way to toot their own horn, to the extent that most people have never heard of the things that Google offers up freely.
Google is not a company like any other.
Re:No matter how innovative (Score:4, Insightful)
Repeat after me: for-profit companies are run by people, not an infallible profit-maximizing robot.
Re: (Score:1)
Of course, if there are no patents, then other people wouldn't know the details of how you did what you did. We are way past the age of simple machines, where you could take apart a typewriter a
Re: (Score:1)
As for space flight, I'm not saying it's easy, but it's easier to answer a question once you've seen the answer. Either way, considering how you feel about pharma patents, wouldn't you agree that if a space company developed a new fuel or insulating material, then they should be able to have some sort of patent on it?
Re: (Score:2)
YOU CAN'T HAVE IT....
Until after the patent expires.
Re: (Score:2)
or you license it, now. I think the real question is when does those patent expires, 20 years for GIFs were ... a tad long indeed
Re: (Score:2)
That's a good point about the licensing. On one hand America is hemorraging jobs overseas. On the other, you have people here on /. whining about Americans wanting to make money on their inventions. It seems to me like either we protect our intellectual property, or we slip further and further behind.
Re: (Score:3, Insightful)
Re: (Score:2)
Nonsense. You don't get a patent just for being the first person to encounter and solve a problem. Your solution has to be non-obvious to somebody else who is skilled in the art.
Re: (Score:2)
As I pointed out, producing energy efficient data centers has become quite a competetive market sector of late, with several quite distinct approaches to the problem, so it's an area that is not exactly lacking for a little R&D either. If the solution was obvious (without the benefit of hindsight) to someone else skilled in the art then that someone else would have done it already and even the USPTO would have rejected
Re: (Score:2)
Even in the most highly competitive marketplace, somebody will always invent something first. However, we don't want to grant patents just for being novel or new, and instead also require that the invention is truly innovative or non-obvious.
Obviously, anybody who invents something first did something "innovative". However, when discussing whether the invention should be protected by a patent monopoly, there's a higher bar for that word than just being the first. Put another way, you're saying that being fi
Re: (Score:2)
We don't even know if they're going to charge for it, maybe they just want to cover they asses from patent trolls. I know I would.
Re: (Score:3, Insightful)
Re: (Score:2)
Patents are not stupid. They are there for anyone to use. If you have great ideas for an invention, why not patent them? Believe it or not, the nuclear power aircraft patent belongs to Dick Feynman. Yeah but you do gotta love paperwork. Invent a better way to submit patent submission and patent that! We'll share the billions.
Re: (Score:2)
What? This patent keeps the planet from from being destroyed? I need to read those claims more carefully. Did I miss a big deflector shield or something? Although, if SciFi/SyFy Channel Original Movies have taught me anything, it's that the majority of planetary threats can be averted by nukes, and I know I didn't see any references to nukes in there.
Oh.... you meant an innovative way to save energy and possibly the humans, or at least their current way of life. Never mind. That's a shame though. I was sta
Re:Does this pass the "Evil" smell test? (Score:4, Insightful)
Hmmm... Is creating patents for things like this "evil"? Seeking to prevent others from saving energy (unless they pay a toll) is not good for this planet, and I'm not sure if passes for "good".
Just pretend they never had the idea at all and nothing has changed. There, feel better?
I know it's popular to bash Google around these parts, but you're really reaching.
Re: (Score:2)
Well, I'm not really bashing. I'm just asking. I actually like Google. But to simply patent the idea of point cooling seems to be stretching it a bit.
If IBM did this, we'd all understand - they are in the hardware business. But for Google to go to the trouble to patent it, seems odd. I'm trying to understand their motivation, and if it is simply to force Yahoo and others to spend more money for power, then I'm not sure it passes the smell test.
Think of how much power data centers all over the world use,
Re: (Score:2)
Also, they're not stopping anyone from using it. What they're doing by patenting, which just means that those that want to use it can pay a licensing fee, or wait until the patent runs out.
Re:Does this pass the "Evil" smell test? (Score:5, Insightful)
That's pretty much the only thing you can do with a patent, stop someone else from using it. (Licensing is just an agreement not to exercise that power.)
I see several explanations for Google applying for a patent:
I wouldn't consider any of these particularly evil, but it is inconvenient for smaller organizations who might want to use the technique, but don't want to go through the hassle of negotiating with Google (who might just ignore their request for licensing).
Re: (Score:3, Insightful)
Re: (Score:2)
That's pretty much the only thing you can do with a patent, stop someone else from using it. (Licensing is just an agreement not to exercise that power.)
You can also stop someone else from patenting your idea and stopping you from using it.
Re: (Score:3, Insightful)
IBM is not primarily in the hardware business... you know that, right? Services are the biggest revenue generator at IBM.
As for Google, what kind of company do you think they are? What's their main gig? Data.
Re: (Score:2)
Considering that so much of their business revolves around data processing, wouldn't you think that a method that reduces one of their largest costs (datacenter operation) is key to their business? If anything, datacenter technology is at the forefront of what Google does.
I think it's strategic protection. Patents work two ways - it's not just protecting against copyists, it's protecting yourself from others as well. If others see an innovation you are using, they could turn around and patent it themselves, then stifle your use of the innovation.
Since Google specify their own server layouts, I can imagine this patent is key to them not having someone else lock them out of a way to make their data centres work efficiently.
Re: (Score:2)
I think it's strategic protection. Patents work two ways - it's not just protecting against copyists, it's protecting yourself from others as well. If others see an innovation you are using, they could turn around and patent it themselves, then stifle your use of the innovation.
Actually, if you're using an innovation already that would constitute prior art and would prevent someone else from getting the patent, or at the least would serve as an effective defence against a patent infringement claim. Which isn't to say that patents can't be used defensively - having a patent could help defend against an infringement claim on a very similar patent. Also, having the relevant patent might let you avoid suit altogether, and avoiding a lawsuit is generally better than winning a suit.
I
Re: (Score:2)
If IBM did this, we'd all understand - they are in the hardware business.
And Google isn't in the data center business? Besides, the idea of point cooling may be obvious, but what they patented is the system of 'air wands' (whatever those are) to accomplish it.
Re: (Score:2)
It looks like a tube with holes punched in one side like those used for duct smoke detector air sampling.
Re: (Score:3, Insightful)
You mean like the rubber hose I used in a system five years ago to force incoming cool air to pass directly over the CPU first and then get sucked out of the box by negative pressure? Wow, didn't know that was patentable. It seemed so ... obvious.
Re: (Score:2)
You mean like the rubber hose I used in a system five years ago to force incoming cool air to pass directly over the CPU
Is that what the patent is for or are you just speculating?
That is a bunk argument. (Score:2)
Just pretend they never had the idea at all and nothing has changed.
Bullshit. That is assuming that no-one else would have had this idea if Google didn't. Considering that practically every major invention has been developed independently by several people, that is a ridiculous position to take.
Patents can absolutely create a situation that is worse off than if the "inventor" had never thought of the idea. Whether Google is "Evil" or not depends on how they choose to use the patent.
Re: (Score:2)
Hmmm... Is creating patents for things like this "evil"? Seeking to prevent others from saving energy (unless they pay a toll) is not good for this planet, and I'm not sure if passes for "good".
Just pretend they never had the idea at all and nothing has changed. There, feel better?
Worse, the poster misunderstood the nature of patents. Patents don't prevent anything. They only give you the legal right to a claim against someone else.
If I get a patent for my invention (say, an egg-cracking lawnmower) in so doing, I'm not preventing anybody else from making an egg-cracking lawnmower. However, if somebody else starts making egg-cracking lawnmowers, then my patent gives me a supportable right to sue the other company. I don't have to do anything, and if I don't, it's no different than not
Re: (Score:2)
The patent doesnt stop some other company from building and using a device exactly like this internally for thier own use. It only stops them from taking the idea and selling a product based on it.
Re: (Score:3, Informative)
The patent doesnt stop some other company from building and using a device exactly like this internally for thier own use. It only stops them from taking the idea and selling a product based on it.
Yes, it does. There is no such thing as "fair use" in patent law. The owner of a patent has the right to exclude others from making, using, importing, selling, or offering to sell the invention. Those first two stop another company from building and using a device exactly like this internally.
And yes, I am a patent agent.
Re: (Score:2)
Hmm, seems you are right...
our patent system is really screwed up, isnt it?
Re: (Score:3, Informative)
Hmm, seems you are right...
our patent system is really screwed up, isnt it?
Why? Wouldn't it be more screwed up to say "gosh, Mr. inventor, I don't have to pay you for a license, because I'm going to use your invention for my own use"? It would prevent them from selling it - so consumer goods for sale would be subject to patents, but manufacturing machines and assembly line innovations that companies install and then use for 20 years would essentially be up for grabs, license-free. No one would ever invent anything for 'behind the scenes' use unless they're employed by a big compan
Re: (Score:2)
Re: (Score:1)
wow, i really messed up my reply there.
I'm honestly interested in how this whole process would work.
Re:Does this pass the "Evil" smell test? (Score:4, Insightful)
As a person who is handy carving things out of wood who might happen to one day create a sitting device for my own person use that is identical to your stool, would I be expected to have to pay you, even if i had never seen your stool?
I don't see how, as an individual, I would not be allowed to make for my own use anything i am physically capable of making with my own materials and tools.
Because this is a civilization rather than a libertarian anarchy? Why should you be able to enjoy the fruits of my labor and sweat of my inventive brow without paying me a reasonable license fee?
As for "never seen my stool", there is such a thing as innocent infringement... For one, you can't be liable for indirect infringement without intention to infringe. For two, damages tend to be lower: if you're truly innocent, I'd get the cost of one stool out of you.
Re: (Score:2)
Why would you expect to get anything from me when we never met, I never saw your invention, I used my own materials, my own "labor and sweat", to create something for my own person use? I could understand if I tried to then sell my stool to others, but barring personal use just blows my mind.
I'd at least hope to have the cost of the phys
Re: (Score:2)
I could understand if I tried to then sell my stool to others
I won't be buying your poop, nor that of anyone else!
Re: (Score:2)
problems with the patent system (Score:2)
I think the patent system is screwed up, but not based on anything in your description. Namely, the patent process strongly favors those with deep pockets, programmers waste a lot of time just trying to figure out what is and is not patented, a small company (and even some large ones) can be sued out of existence just for infringing a patent they neve
Re: (Score:2)
apparently I was wrong about this. I will now commit rutual seppuku to prevent the savage flaming i am no doubt abount to receive.
Re: (Score:3, Funny)
Death is no escape from a savage flaming on slashdot. It just means we will be crisping your corpse and be denied the joys of imagining the sounds of your lamentations and woe. On second thought, we could just *pretend* you didn't commit seppuku, and still have those joys. So really, your effort goes wasted -- please put the wakizashi down and resume whatever it was you were doing.
On the other hand, maybe you
Re: (Score:1)
Hmmm... Is creating patents for things like this "evil"? Seeking to prevent others from saving energy (unless they pay a toll) is not good for this planet, and I'm not sure if passes for "good".
A patent merely grants you control over the invention's licensing, it is possible to wield a patent in an open manner; or not. We'll have to see how they play it.
However, unlike some of the more obvious software patents floating around; this sounds like a genuinely clever idea. If they choose to sit on it to stay ahead of their competition; I don't know that it's fair to say they're "preventing" anyone from saving energy. It's not like anyone would have used this method had they not invented it; and t
Re: (Score:1)
Why? (Score:2)
As it is, I think that the approach that they are taking is actually the wrong way. OTH, It is better than what is currently done. Others like MS could license it. More likely MS will introduce a similar patent and it will go through due to how stupid the US patent office has become.
Re: (Score:2, Interesting)
Re: (Score:2)
Re: (Score:2)
If they don't patent it, someone else will. If they allow everbody to use the patent, then it is still not evil, even if they charge a nominal fee to do so.
Once you publicly show the invention, someone else can't patent it. Well, it is supposed to work that way.
Soon (Score:5, Funny)
There have been 17000 youtube searches for Natalie Portman naked in the last 450 milliseconds. We need a burst of cold air on rack 1000001, processor 304 on the second chip on the third stick of RAM.
Re:Soon (Score:4, Funny)
At least we know the DB servers are cool (Score:3, Funny)
They can cross that off the list and carry on with other solutions to the
Error establishing a database connection
problem...
Re:At least we know the DB servers are cool (Score:5, Funny)
Obvious patent (Score:1)
Going off the summary...
I invented that! Money please!
For real though, if people listened to me once in a while they would be rich. How do I cut out the middle man.....
Patenting the obvious (again) (Score:2, Insightful)
Patent Attorney: Great idea! How many different ways can we cool it? When can I have a New Invention Report?
I can only hope that this straw contributes to breaking the camel's back.
Sweet.. sweet... (Score:3, Funny)
Error establishing a database connection (Score:1)
Don't tell me "Datacenter Knowledge" got slashdotted.....not good for teh community.
Not a Patent (Score:2, Insightful)
Seriously - is it too hard to note that it's only a Patent Application at this point?
Re: (Score:2)
Seriously - is it too hard to note that it's only a Patent Application at this point?
Do you really believe that Google, with an entire department of some of the best IP attorneys in the world, has a very high rejection rate for their patent applications? Seriously?!
Same heat flow (Score:3, Interesting)
Re: (Score:1, Interesting)
The way I see it working is that by providing focused cooling where it is needed most, the rest can be sufficiently cooled at a higher ambient temperature, thereby reducing the energy needed to remove the same amout of waste heat.
Re:Same heat flow (Score:4, Insightful)
You are right - you need to remove exactly as much heat as the equipment is generating. The energy savings with this scheme is due to the fact efficiency of chillers is lowest when asked to produce coldest output. Traditional data centers keep the hot parts cold by keeping everything very cold. Efficiency is improved if you can run your chiller at a higher output temperature and compensate for the reduced effectiveness of the warmer air by directing it where it is most needed.
Re: (Score:2)
In practice this depends entirely on the design of the chiller - specifically the choice of working fluid and the choice of pressures. The refrigerator in your kitchen is happy to efficiently exhaust heat from freezing air to sweltering air. Heat pumps' theoretical limits of efficiency are the same as heat engines in reverse. The carnot efficiency of an engine with a 290K cold side a
Re: (Score:2)
The cooling design for data centres needs to shift from an office cooling design to an industrial design. Once you get to that size data units should be shifted to auto-pick racks so you can run the whole facility beyond human comfort levels in the mid 30s centigrade. That way you only pre-condition fresh air and then exhaust rather than recycle. This allows you to adjust your cooling air from 100% fresh, to ground water cooling, to evaporative to refrigerated(ammonia) as external ambient conditions and he
Big whoop. (Score:2)
First, having a patent means basically nothing. It doesn't mean it works, that it's practical, or that it will be used by anyone.
Second, this type of patent is going to be very difficult, if not impossible, to enforce. Prior art aside, how many data centers do you know that are open to public inspection? They can't just go buy a widget to compare them.
Seems like little more than a trophy (framed patent) for someone's office wall.
Directed cooling. (Score:2)
Military avionics has had very directed cooling for years. Cool air is a scarce resource in field systems. Here's an engineered cooling system [vmecritical.com] for VME boards. Military PC boards tend to have covers over the components. ("When someone is fixing your system, it's too hot, too cold, too dusty, or too wet, and someone may be shooting at them." - a reminder given military hardware designers.) Those covers can be designed to direct airflow. The covers then plug into an air plenum which feeds air into the cov
Power Supplies (Score:2)