Tuesday, 3 September 2013

SQL table design strategy for large amounts of location data with lots of inserts

SQL table design strategy for large amounts of location data with lots of
inserts

I'm designing an ASP.NET MVC web application that will be used to find
where people are and map where they go from different apps. The system has
the potential to have a large amount of location data for every user,
about 3 months back for each user. I'm now trying to decide how to design
a MS SQL 2008/2012 database for this.
Naively, there would be a table of location updates with following columns:
Id (int)
UserId (int)
Latitude (double)
Longitude (double)
Speed (int)
DateSent (datetime)
From my reading, it seems partitioning the table is the best way to
accomplish this. If that's true, I'm a little confused on how it becomes
automated to say divide the data in to partitions such as "today", "this
week", "last week", "previous month", etc all the way down to having one
large archive of everything that is just used for backup.
If I can provide any more technical requirements that might be helpful is
devising a strategy let me know. I appreciate any insights from some
database experts out there.

No comments:

Post a Comment