February 6, 2008 at 11:26 am
I am doing some research on Partitioning and I came across this title during a search. I would like to find out more about how this can be accomplished from an active table. I am trying to minimize the down time to the site while the data is being partitioned.
I will need to partition the data on a computed column so that the data will be distributed evenly on ongoing basis. I will take a key value and divide by an integer to form the partitionid.
The source table is over 500 million rows and the client does not want to consider archiving at this time. Can anyone shed some light on this?
🙂
February 6, 2008 at 1:22 pm
Here is my cent.
Using a cursor to partition by parts continuously or schedule a job to run periodically until it is done.
February 7, 2008 at 3:51 am
I had a similar task not so long ago.
not sure if its the right way, but it worked for me.
I Added a field into the table called LastModified with the last modified date in it.
I copied the table structure and created the following tables.
tbl_Weekly
tbl_monthly
tbl_Quarterly
tbl_Yearly
Then created a view with all these tables joined in union.
Finally, I created a script that would look at each tables LastModifed date and place the record into the appropriate table(s) based on todays date. This script runs once a night.
so far, its working great for record return speed.
I am only a rookie thou, so if anyone sees a flaw in it.. let me know. 🙂
Viewing 3 posts - 1 through 2 (of 2 total)
You must be logged in to reply to this topic. Login to reply