Wayne Eckerson at TDWI wrote a nice overview and checklist about doing analytics on large data sets.  Note this report was sponsored by Aster Data but it is a nice overview and not a sales brochure.  Wayne explains what big data is – "data volumes are exploding" and "it is now time to start a petabyte club, since a handful of companies, including Internet businesses, banks, and telecommunications companies, have publicly announced that their data warehouses will soon exceed a petabyte of data."  Wayne then talks about "deep analytics" which "ranges from statistics—such as moving averages, correlations, and regressions—to complex functions such as graph analysis, market basket analysis, and tokenization."  Then there is a nice checklist of what to do when faced with doing analytics on big data:

1. Educate your users

2. Determine the type of analytics you need

3. Architect appropriately for big data analytics

4. Use "in-database" analytics

5. Don't limit your analytics to SQL – Wayne talks about MapReduce, but there is more than just MapReduce here

6. Define your requirements before selecting the right product

This is a quick read and worth a few minutes: http://www.asterdata.com/resources/assets/ar_TDWI_ChecklistReport_BigDataAnalytics.pdf