June 2, 2015 at 8:01 am
I work primarily with the data science wing where the data I manage is used as such.
Most of the SQL I write is to prep for datasets that are going to be pushed into a statistical environment such as R to do the final analyze that possibly could push back into SQL. The goal is market attribution, which is analyzing hundreds-of-millions of records to paint a holistic picture of the end user from browsing to purchase. Everything in between is computed, crunched, analyzed and consumed with statistics and everything else.
Due to that, most of my function use is restricted to the basics because when you're dealing with that level of computing, we sometimes depend on other services to help with the load rather than slamming it all into SQL Server directly. This may change in the future with Microsofts new offerings.
Viewing post 16 (of 16 total)
You must be logged in to reply to this topic. Login to reply