Viewing 15 posts - 51,826 through 51,840 (of 59,072 total)
Grant Fritchey (3/12/2008)
It sounds like a paging problem.One, of many, solutions that Itzik Ben-Gan offers in TSQL Querying:
I wish great authors would warn people of the code the create. ...
March 12, 2008 at 5:35 pm
I built a million row table as you described. I made it so that the Status column has only 5 different possiblities and verified the distribution as an average...
March 12, 2008 at 5:24 pm
Adding columns to a temp table or any table that uses ALTER is always a PITA... during run time, the new columns are perceived as NOT THERE. That means...
March 12, 2008 at 3:03 pm
It simply substracts 2 from the character position returned by the FINDSTRING.
March 12, 2008 at 2:34 pm
madeley68 (3/12/2008)
March 12, 2008 at 12:01 pm
Now there's an idea... and properly partitioned, it would be updateable, as well.
March 12, 2008 at 11:47 am
Perfect... thanks for the feedback!
March 12, 2008 at 11:43 am
Lynn Pettis (3/11/2008)
March 12, 2008 at 11:10 am
Great link for a "System DBA", Adam. What about an "Application DBA" or a "Hybrid"?
March 12, 2008 at 8:37 am
Isn't a DB2 timestamp is listed in seconds since some date? Are we chasing a ghost just because the OP said "it used to work"?
March 12, 2008 at 8:31 am
Heh... and, still, you have to pay attention to the title... "ESTIMATING the size of a clustered index."
March 12, 2008 at 8:27 am
Brandie Tarvin (3/12/2008)
March 12, 2008 at 8:23 am
Matt Miller (3/12/2008)
March 12, 2008 at 8:07 am
Then, pick one of the two methods in the code I demo'd above...
March 12, 2008 at 8:03 am
Sorry... the table is pretty much toast, then.
March 12, 2008 at 8:02 am
Viewing 15 posts - 51,826 through 51,840 (of 59,072 total)