How Google Works
Before you can understand how Google works, you probably should
have a basic idea of how the Web works. When you visit a website, your browser
is actually contacting a web server, a computer whose job is to
deliver web pages. So when you click a link, your browser contacts the server
and says, "Send me this page." The server takes the request and then sends the
page to the browser, which displays it on your computer.
Key Term
Server A computer whose
job is to perform a specialized task and deliver information. For example, a web server serves up websites, while an email server sends or
receives email.
Okay, now that you have that basic background down, let's see
how Google works.
In some very basic ways, Google works just like other search
engines. Its basic operations
are exactly the same. Like all search engines, Google is composed of three
parts:
Key Term
Search engine A site that
allows you to search the Web.
The Spider
The spider part of the Google search engine is an automated
piece of software, also called a robot, that requests many thousands of pages from
hundreds of websites simultaneously. When it finds links on
pages, it follows those, and requests those as well.
The main Google spider is the GoogleBot, and it essentially crawls the Web once a
month. Obviously, many sites change more than once a month, and so Google also
has a crawler named FreshBot that crawls pages
constantly.
The Indexer
The spiders send information about all
the pages they find to the indexer part of the search
engine. The indexer then does a pretty
amazing jobit creates an index of every word on every page sent to it by the
Google spider. Not only does it index every word and every URL, it also keeps a
record of where every word is on every page.
Multiple copies of this index are kept on various Google servers. A single server wouldn't be able to keep up with all the
search requests that are done.
The Query Engine
The only part of Google that you see is the query engine, and
you only see part of that. It's the public face
of Googlethat inviting search box at the top of Google pages.
When you type a search term, a Google web server sends your
request to the indexer, which is housed on multiple indexing servers. The index
servers look through the index and match what they find with your request. The
index server then sends that information to document servers, which retrieve the
correct information and format it so your browser can understand it. That
formatted information is then sent to your browser.
And it all happens in a fraction of a second.
Google's Special Sauce
All this search engine logic is nothing new or revolutionary.
This technology has been around for years, long before Google was a glimmer in
its founders' eyes.
So why is Google so good at what it does?
Google uses better algorithms than any other search
engine, and constantly refines them. Algorithms are sets of
rules for performing a particular task. In Google's case, its algorithms are
responsible for taking your search request and deciding which results to show
you.
Algorithm A set of rules
for performing a task. In Google's case, algorithms are what determines which
pages it says match your search requests.
Google's algorithms aren't particularly easy for mere mortals
to understand, they're changing all the time, and they're not made public.
Google uses more than 100 factors in its algorithms. For every search you do, it
considers all of those factors and then calculates a score for every possible
matching page. The page with the highest score is the first search result. The
page with the second-highest score is the second search result, and so on.