Clickstream

Google Paper on Querying Massive Amounts of Data

Google labs has an interesting paper on a method for going through large volumes of data. New techniques are needed for dealing with extremely large data warehouses. I've been looking at a couple of tens to a hundred terabyte data warehouses and trying to figure out how to address the problem of detailed questions. This paper describes one possibility, assuming that you have a lot of computers available.

Also a nice animated map of google search data for a 24 hour period in 2003, in case you haven't had enough of animated maps postings yet.
link to animation
link to paper

Posted by Mark Sunday, January 15, 2006 6:38:00 PM | permalink |