November 10, 2013 at 12:40 am
Hi all,
I was reading up on this article:
http://www.sqlservercentral.com/articles/Integration+Services+(SSIS)/102924/
And I was wondering in the article it was about updating the tables of your database but not a complete new insert. Once a day a import runs to import every table (again). I was wondering is option 2 mentioned in the article not better then one single import run??
November 10, 2013 at 1:47 am
prennings (11/10/2013)
Hi all,I was reading up on this article:
http://www.sqlservercentral.com/articles/Integration+Services+(SSIS)/102924/
And I was wondering in the article it was about updating the tables of your database but not a complete new insert. Once a day a import runs to import every table (again). I was wondering is option 2 mentioned in the article not better then one single import run??
Here [/url]is your link again, this time in click-able form.
The gist of the article is that row-by-row updates using the OLEDBCOMMAND component are slow, but running such updates in parallel is less slow. It's definitely not fast, whatever you do.
Finding a set-based solution usually leads to the best performance (though you may need to consider batching updates and deletes to avoid excess tlog pressure).
Back to your question, which is a little unclear, to be honest. You seem to be asking whether it is better to UPDATE than to INSERT? I don't see how you can replace one with the other, unless your INSERT is preceded by a DELETE, perhaps.
What do you mean by 'single import run'?
The absence of evidence is not evidence of absence
- Martin Rees
The absence of consumable DDL, sample data and desired results is, however, evidence of the absence of my response
- Phil Parkin
Viewing 2 posts - 1 through 1 (of 1 total)
You must be logged in to reply to this topic. Login to reply