June 6, 2015 at 6:52 pm
Also, to continue, there is another discussion that seems to expand on the OVER (ORDER BY ...) technique, adding that using function ROW_NUMBER on large tables can be counterproductive. There are multiple solutions in the replies to this blog article:
http://stackoverflow.com/questions/15815624/sql-performance-using-row-number-and-dynamic-order-by
June 8, 2015 at 7:33 am
Good article. I often use case in order by. This will give me some new ammunition on how to use it better.
June 8, 2015 at 9:01 am
My basic question on all this is: yes, but WHY does one need it ?
Maybe building up an additional field such as a running subtotal ? For a report ?
My own purpose for dynamic ordering was to allow users to determine the sorting order of a list of invoices to be manually selected for printing. From the SSRS/Report Builder User Interface. Speed and performance of presenting an ordered list in such a case are not really a consideration (a one or two second response time should be adequate and no one really wants to view a list of 1,000,000 items ...). Since new invoices can appear during the time the user is selecting older ones for printing, READ UNCOMMITTED should remain a viable option which should not (I presume, lurching totally above my level of knowledge here) cause blocking.
So, under which other circumstances would there be a need for high-volume behind the scenes process based on ordering a list ?
And by the way, good call on Brent Ozar's great link.
June 10, 2015 at 8:13 am
Ben, I'd like to see you write an article that shows more of what you want/mean about LINQ vs TSQL. I don't know if you're correct or not, but I'd like to see it illustrated.
Viewing 4 posts - 61 through 63 (of 63 total)
You must be logged in to reply to this topic. Login to reply