Slashdot Mirror
How does Google do it?
Doc Tagle writes "With Google reportedly on the verge of going public, more and more people want to know what makes Google tick. The Observer, serves up the answers to our questions."
← Back to Stories (view on slashdot.org)
If truth is the first casualty of war, openness is the first casualty of going public
OK - I can (perhaps) see this as being the case prior to an IPO, but that statement can't be true after it has happened...
I mean....surely once they've gone public, they'll be obliged to detail and list the sort of information that the article postulates about? The shareholders would be entitled to know how many servers google has, what their specifications are, and what their current commercial strategy is.....surely?!
Google has been at 4.285 billion pages for more than three months straight. The count hasn't increased in a long time... The index is maxed.
Google has recently removed tens of thousands of "duplicate content" sites from its index - where "duplicate content" is as simple as being an affiliate site (e.g. Amazon) and having the same textual item descriptions as many other sites.
Google is now in the process of dropping millions of link records from its index, presumably to make room for more pages.
Google is wavering.
Gmail is a distraction, a venture into some other space to keep people from noticing that their search product is degrading.
May she last as long as possible...
PigeonRank! Duhhhhhh
> If truth is the first casualty of war, openness is the first casualty of going public.
Maybe this is the reason after all, but I think it's more about Google being simple, smart and clean. They play fair (no browser interstitials, no sneaky crap, no registration necessary...etc); I would equate Google's victory thusfar to a kind of no-nonsense attitude to business, always, no-exception.
The dangers of knowledge trigger emotional distress in human beings.
I read the article and it didn't say much at all about how Google operated. Instead, it just said we don't know how they operate because they keep it secret. But maybe that was the point to begin with.
-Vic
The only thing it's missing now (IMO) is spellcheck and an online translator, which I'm sure they're already planning. I'm also looking forward to Gmail being open to the public. After they conquer these 3 thing, whats next.. Google ISP? Google National Army?
having been a consultant at their data center a year or so back I can attest that they had well over 50,000 machines. I am not sure about the 80GB drive per machine because from what I understood was they bought whatever drive at the time was the cheapest MB/$ and would replace any dead ones with the larger ones. Also, at any given time machines just die and many of them are not replaced or repaird for months. Their cluster accounts for all this...
-eric
I lost a couple of sites from Google this month, presumably due to duplicate content; they were nearly verbatim clones of some of my other sites. The original sites are still there, the "clones" vanished from Google. As in, even if I search for those domains directly, I get nothing, where I used to get a cached copy of the sites. They've quite literally vanished from Google's database.
Can you back up your assertions that Google's index is full? It's a rather interesting theory, and perhaps an explanation for all the tweaking they've done lately.
"BSD: Free as in speech. Linux: Free as in beer. Windows 10: Free as in herpes." --Man On Pink Corner in #52607549.
One -- Slashdot seems to be into content-directed ads now... as google was my ad for this story.
Two -- If you want your pages indexed faster and more frequently, sign-up and place a google adsense ad on your page. Many webmasters believe that google is having to index so many adsense pages... that is difficult for google to add many more non-ad driven pages.
Just sign up for adsense and run it a couple of weeks while you build your site. After google has spidered your site well, then just drop adsense.
Good luck. I would love to hear any of your google-related tricks.
AC
They will not have to disclose the number of machines, the OS, the anything related to the machines. Wall Street isn't buying their technology, they are buying their cash flow.
If you do not believe me, buy a share of GE. Pick up the phone, call Investor Relations and ask them how many Unix computers they have and what OS and patch level they run.
I mean....surely once they've gone public, they'll be obliged to detail and list the sort of information that the article postulates about? The shareholders would be entitled to know how many servers google has, what their specifications are, and what their current commercial strategy is.....surely?!
Why would a shareholder care about server specifications? Investing is all about money. Read any quarterly report from a public company. Income statement, balance sheet, and cash flow are the primary interests on the numbers side as well as a general roadmap of where the company's heading. Warren Buffett doesn't care if each server has two 80 GB drives, or whether they have four 250 GB drives per server. The only thing that matters is that there are competent people to handle these kinds of "dirty details" that an investor doesn't give a rats ass about.
Take a look at the kinds of information you could expect from Google's quarterly reports.
Want to improve your Karma? Instead of "Post Anonymously", try the "Post Humously" option.
On the other hand, here's the conspiracy theory version: what if Google IS the NSA? The IPO is a smokescreen to try to avert attention. The reason they can't show their true capability is that when the company goes public, only 20% of their hardware will actually go into the public company "Google", the rest of the hardware will still be hidden and a part of the NSA's system. :-)
[For the humor impaired, I'm just joking, but it does make you wonder...]
Craig Steffen
http://www.craigsteffen.net
Putting on my computer scientist hat I would guess:
- instead of backup, hold data in multiple places at once
- use a "cascaded rsync" to trickle software changes to thousands of nodes
- then load software via NFS at node bootup
- use nodes just to store data; keep software in RAM for speed
Just a few thoughts.
GIMMEE would be nice. Well, nice for awhile and if they didn't get weird with it. Don't know if that could happen though, nature of man and all that philosophical stuff. Goes along with the current VoIP articles. They would dominate the net then if they implemented that. I know I would pay cash to them have a universal works great, any OS VoIP and no-spam, no commercial email service.
So far we know they have just a cubic load of servers, the most on the planet most likely with one private company. The government probably has more, but it's a mish mash of them, not near as sleek or coordinated, AFAIK. What COULD be next with them, practical cheap 50 dollar thin clinets that you could do a TON on, using distributed computing, from games to communication to running any business? With tech savvy like they got and their already established heavy hardware base and heavy committment to R&D, they could just 'splode with an extra 25 billion in cash all of a sudden from an IPO. OR, the money could get to them and they become just another weird company that forgets it's roots as "brains come first" and switch to "marketing crap comes first" like certain other unnamed megacorps do now.
Interesting times
For those who haven't read - there is an article written by Brin and Page - maybe a little outdated, but still interesting: The Anatomy of a Large-Scale Hypertextual Web Search Engine
There is plenty of evidence to suggest that Google has run out of docid's, hitting the 32-bit integer limit.
The best evidence is doing a search which returns results which say "Supplemental Result" next to them. That'll be coming from a second document store I'd guess.
I disagree. An investor deserves to know at least general information about the goings on of a business. If I were a stock broker I would want to know that say: FruitCompanyA uses insecticide whereas FruitCompanyB doesn't. I personally would choose FruitCompanyA as a a rise in the insect population would ruin FruitCompanyB.
With google: before I give them my money, I would like to know how many servers they have, how close to capacity they are, what softwares they use (compatibility issues).
Honest reporting of operations lets an investor make an intelligent decision about their money and helps avoid boiler-room companies.
unfortunately the technology spending IS part of the cash flow. "We went dumpster-diving and picked up a dozen new machines for the indexing farm" and "we entered agreement with Dell to secure a reliable source of cheap Intel servers" would both show up on the shareholder statements but the impact would not be the same.
Going public WILL expose the siginificant portion of Google technology, more sp when it has to do with hardware.
The problem with that analogy is that what software they run has absolutely nothing to do with what they do to make money.
With Google, their entire "business" - their means of generating cash flow - relies on sheer quantity of computing muscle and high performance software for their search databases. With GE, their business is making lightbulbs, dishwashers, hair dryers, electric motors and any more of thousands of different products used in residential, commercial and industrial settings. How many Unix computers they have in all their offices around the world is a causality of doing business, not their means of doing business.
I'm sure if you asked the GE Investor Relations department something relevant about how their business operates, you might get somewhere.
=Smidge=
Recycling without attribution is the first casualty of bad journalism.
I thought I had read this article before, and then I realised, I had read it before...
(although I now realise that you are not supposed to read the linked articles before posting comments - sorry)
Humorous signatures are over-rated.
Google search for the letter "a" resulted in 3,530,000,000 hits [search took 0.12 seconds].
Neat. I wonder what doing a Google search would return for other letters:
"c" -- 299,792,458 hits
"e" -- 2.71828183 hits
"h" -- 6.626068 × 10^-34 hits
"i" -- sqrt(-1) hits
"k" -- 1.3806503 × 10^-23 hits
Looks like Google is definitely busted. They should fix these bugs.
Want to improve your Karma? Instead of "Post Anonymously", try the "Post Humously" option.
With google: before I give them my money, I would like to know how many servers they have, how close to capacity they are, what softwares they use (compatibility issues).
I agree it would be nice to know. But if those are your conditions for investing in Google, I think Google would probably tell you to keep your money. I imagine Google's quarterly reports would probably say something like:
"Our operation depends on having the ability to increase our server and bandwidth resources as we grow our services. Business may be adversely impacted should capacity be unavailable. Our servers are also at risk for viruses, worms, and DDoS attacks which could put the operation of those servers at risk and adversely affect business." etc...
That would give you, as an investor, the information you need to determine whether those risks are worth your money. In all likelihood you'll just have to rely on the fact that they have an army of PhDs who are smarter than you and I put together and know their shit when it comes to security, databases, clustering, etc.
Now I could be wrong. Perhaps Google is waiting for the IPO and will then detail their server infrastructure, wow Wall Street (and geeks worldwide) with their amazing capacity, and their stock will skyrocket on the first day of trading. I'd wager that Google's stock is going to have amazing gains anyway given that it's a bit of an industry darling. Other tech companies which have been thinking of going public would be wise to time their IPO very shortly after Google's and ride the wave.
Want to improve your Karma? Instead of "Post Anonymously", try the "Post Humously" option.
> 1) Why are their terms of service / Pirvacy Policy so vague?
This is to keep it simple. Exacting legal language is the path to screwing people. Vague terms of service are good because both sides can wiggle. Has anyone been sued because of these terms of service? I'd like to see some refs to that, but I'm guessing it's just to protect the general public from a-holes who would exploit Google.
> 2) Why does their cookie stay until the year 2038?
Not to be funny, but someone at Google likely knows when the end of the world is coming and has set the cookie to reflect this. Seriously, who cares how long cookies stay alive for? You can block them if you like, but I think it's really just to keep Google more effective.
> 3) Why does their Google search bar report information and auto-update without permission?
I'm against Spyware, so I don't run it, but Google tracks searches anyway, so what's the point of getting upset about it? These technologies makes Google more user-friendly. Google doesn't have loads of popups trying to get you to install the bar -- it's not right in your face. People who want it likely don't care if it auto-updates because then they have the most recent version of it.
The dangers of knowledge trigger emotional distress in human beings.
If that link gets slashdotted, here is another link of a PDF PowerPoint presenation.
Good read! This paper (with the discusion of the goodness/fastness of file appends) made me more interested in Prevalence - so much so that I am using it for my new project.
-Mark
Another wonderful speculation about Google infrastructure which You can find it here.
As far as I can tell there is no better way for that hardware to have come "back into the community."
The service is free, and they're really good at what they do. I would say I'd be lost without google on the internet, but really this compliment goes for lots of search engines - I'm really very grateful this sort of service still exists for free (well, with ads.)
Unless you want to talk about cures for diseases through protien folding simulations, I can't think of a better way for this hardware to be used, such that it begets a greater net benefit.
The snow doesn't give a soft white damn whom it touches. -- ee cummings
Robot.txt
The Google bot respects it, so if you're up to no good, it's easy to get Google to not index your page.
Anyway, I'd like to see a version of google that didn't respect robot.txt. You'd used to be able to dig up alot of infermation on peopel on google before they started to use robot.txt on alot of sites.
Sometimes I wish I was a plumber, then I'd know how to deal with other people's shit.
"To be or not to be"
and I honestly can't see what you are going on about: of the first ten results, eight highlighted the phrase in the page synopsis, one used the phrase as a domain name, and one included the parital phrase "...Or Not To Be."
Note the elipsis on that last one: it alludes to a larger portion of text preceding the printed portion. And the domain-name was found even though the spaces were omitted.
Those aren't irregular results: those are highly intelligent results.
Just because they aren't deterministic enough for you to plug them into a piece of code of your own construction (without compensating Google) doesn't mean that they don't fulfill the purpose of the web search.
What is the difference between a small revolutionary change and a large evolutionary change?
The original analogy is a little off. However, if you look at eBay, do they disclose how many systems they are running? How about Amazon? Do I care?
The real fact of the matter is, they have custom software that they run. The number of systems, speed, memory and OSs are simply a byproduct of what they really offer: a service.
Google is no different. They offer a service. As long as they are profitable, as an investor, I could care less if the systems were running on Dell's, White Boxes, Mac, or Commodore-64s. They have found a way to make the business run on the systems they have.
Come on, the nodes in their clusters are not desktop computers with office software on it.
The system running these machines are rather very stipped down: They only need very few applications and a very simple kernel (not many device drivers, maybe no graphic card driver, ...).
Furthermore there are no local users on the the machines -> many security flaws wont affect the integrity. And remote holes in the kernel occur not very often.
And above all these cluster nodes are certaily shielded by some sort of firewall. Therefore they don't have to care for network security themselves.
All in all: I believe that you need to update such machines rather infrequent. At least not for security reasons.
Titus
The problem is, I've never paid these people a single penny for ANY of this. How the hell are they going to make money?
Um, you do realize that Google already makes a profit, don't you? I daresay the IPO will puff the value of the company up beyond the rational amount, but that's not 'Enron' -- if you are going to use buzzwords, use the right ones. Enron was a case of internal actors in the company using financial games to siphon off profits and inflate the value of the company on the books. You accusing Google of financial fraud? If you are going to use a buzzword, use 'Yahoo' or something -- a solid company that got its stock price puffed up excessively due to investor mania.
How the hell did this get moderated up, except as 'Funny'?
Results 1 - 10 of about 5,750,000,000 for the [definition]. (0.11 seconds)
Doesn't that imply more than 4.285 billion?
This issue is a bit more complicated than you think.
"Google manages to achieve this with sophisticated techniques for rippling changes through the cluster, yet achieves 100 per cent uptime. This is serious stuff, and there are a lot of IT managers out there who would give their eye-teeth to be able to do it half as well."
Sigh...as an IT manager I can only dream of 50% uptime. Damn you, Google!
1) Google has an effective advertisement system
2) My last two employers bought Google boxes for their intranet
Correction, the ad model has proven to be of dubious effectiveness with companies that have no credibility.
Google is perhaps the most trusted company on the net today, and with the traffic they get, I'm not surprised at all that they can support all their financial needs with ad revenue, especially with some of the big bucks that large companies dump into advertising with Google. I challenge you to show evidence showing that their advertising business model cannot support their costs, because so far you've done nothing but toss up tin-foil hat ideas without any proof to back it up, and as someone else so kindly pointed out to you, Google is ALREADY in the black.
Buy Steampunk Clothing Online!