What is Google Dance?

During the month, Google sends out robots (spiders) to crawl the web and archive every website it finds. These archived websites will be presented in the next update of the Google index (database). The update process is initiated every 30 days or so, although on some occasions there have been more than 30 days between updates.

Now if you've been a keen Google fan you'll know that they have 3 main www servers online, which are as follows:

www.google.com - The main address, the true homepage of google.
www2.google.com - Thought to be a test server.
www3.google.com - Thought to be a test server.

Whilst nobody outside of "Google" truly seems to know the purpose of the www2 & www3 servers, many people have come to the assumption that they are test servers. During the update, which takes several days, the 3 Google servers display different results. Whilst the results vary from server to server, they are said to be "dancing", hence the name "Google Dance".

The www2 & www3 servers are the ones most closely observed during the dance, however I have recently found the addresses to all 10 of the Google Datacenters!! I believe that watching this will provide for a slight enhancement in the detection of the "Google Dance".. My tool is the only tool (that I know of) to provide the 10 datacenters option. 

Google Servers Architecture 

We have structured our system so that most accesses to the index and other data structures involved in answering a query are read-only: Updates are relatively infrequent, and we can often perform them safely by diverting queries away from a service replica during an update. This principle sidesteps many of the consistency issues that typically arise in using a general-purpose database.
We also aggressively exploit the very large amounts of inherent parallelism in the application: For example, we transform the lookup
of matching documents in a large index into many lookups for matching documents in a set of smaller indices, followed by a relatively inexpensive merging step. Similarly, we divide the query stream into multiple streams, each handled by a cluster. Adding machines to each pool increases serving capacity, and adding shards accommodates index growth...

Read the full story of Google 15000 servers secrest »

Published by the IEEE Computer Society 0272-1732/03 ©  2003 IEEE  

More information: Knowledge Base - What is a Search Engine