Sandbox Spin

[HHI Golf Guy]

There has been a ton of speculation as to whether the Sandbox effect applies to domain, backlinks, new content, Satan, and many other theories.

What if the answer was more simple than all of that?

Could it be possible that with all of the billions and billions of pages that they have indexed (far more than other SE's) that they do not have the computing power to process all of the pages with such a complex algorithm?

In this scenario, new sites would be indexed and then have their content parsed by a "preliminary" or broad based algo. Over time, these sites would be integrated into the main algorithm across all of the data centers.

Even with a vast network of PC's, I can't imagine how much time it would take to process 8+ billion pages through an algorithm with hundreds (if not thousands) of variables.

[crew]

with 100,000 servers, I don't think Google is having any trouble indexing and analyzing 8 billion pages.

[exam]

I think Google has more computing power than you and I can comprehend. I believe everything they are doing is very deliberate. Read this blog entry (http://blog.topix.net/archives/000016.html) Have you ever seen that many comments?

[ferret77]

what comments?

[exam]

It wouldn't finish loading for me either in IE, but in firefox it does fine. :)

[HHI Golf Guy]

Granted, from the reported information G has alot of computing power. But they would need alot of computing power to process 8 billion pages through a relatively modest algo. If their algo is complex with a ton of variables, it could take a while to process the info for their db even with 100k servers.

For those that are not familiar with the nature of algos, there is alot more to them than algebra. Think of them as thousands of nested and looped if then statements that change parameters based upon the absence, presence, or weight of each variable. A slight change in each variable causes a cascading effect throughout the rest of the algo.

It would also seem to make sense, processing wise, that there are variations or multiple forms of the algo to classify data in their database. A site would need to pass certain gateways (time factors, on page factors, of page factors, etc) in order to move on to the next level in the ranking process. This could ease the burden of the processors as well as explain why some "good" sites are held back. If you missfire on one gateway your site is held back from being listed in the top ranking db.

Anyway, this is all pure speculation. I'm just fascinated by the math and the processing power that it must take to pull together a sorted database of that size.

[exam]

Yes, it would be quite a challenge, but I think it would be fun to work on something like that :)