SQL Server
Copied!
Home
Options
849
SQL-Server-Team on 03-23-2019 11:27 AM
736
SQL-Server-Team on 03-23-2019 11:25 AM
199
Craig_Freedman on 03-23-2019 11:23 AM
192
SQL-Server-Team on 03-23-2019 11:18 AM
142
Craig_Freedman on 03-23-2019 11:18 AM
207
Craig_Freedman on 03-23-2019 11:17 AM
192
Craig_Freedman on 03-23-2019 11:16 AM
306
Craig_Freedman on 03-23-2019 11:14 AM
188
Craig_Freedman on 03-23-2019 11:13 AM
230
Craig_Freedman on 03-23-2019 11:12 AM
164
Craig_Freedman on 03-23-2019 11:04 AM
142
Craig_Freedman on 03-23-2019 11:04 AM
145
Craig_Freedman on 03-23-2019 11:03 AM
170
SQL-Server-Team on 03-23-2019 11:03 AM
131
Craig_Freedman on 03-23-2019 11:01 AM
126
Craig_Freedman on 03-23-2019 11:00 AM
136
Craig_Freedman on 03-23-2019 10:58 AM
206
Craig_Freedman on 03-23-2019 10:57 AM
155
Craig_Freedman on 03-23-2019 05:22 AM
157
Craig_Freedman on 03-23-2019 05:21 AM
106
Craig_Freedman on 03-23-2019 05:20 AM
144
Craig_Freedman on 03-23-2019 05:19 AM
143
Craig_Freedman on 03-23-2019 05:16 AM
131
Craig_Freedman on 03-23-2019 05:14 AM
148
Craig_Freedman on 03-23-2019 05:12 AM
137
Craig_Freedman on 03-23-2019 05:11 AM
124
Craig_Freedman on 03-23-2019 05:10 AM
154
Craig_Freedman on 03-23-2019 05:09 AM
186
Craig_Freedman on 03-23-2019 05:09 AM
160
Craig_Freedman on 03-23-2019 05:08 AM
Latest Comments
Let's hope those customers with many cores are running software from vendors that are advanced enough to implement something like this. I've had discussions with vendors to try to convince them that using cascading updates is a better way to cascade foreign key changes than "instead of" triggers. An...
1 Likes
@jahummer B-trees are self-balancing due to the way they are built from the bottom up (i.e. leaf first, then push up a new level when needed), so they never need to be "re-balanced" per se. Even a sequential index may need to be reorganized, however, if there are a lot of in place updates that cause...
0 Likes
@Rmwin Thanks for the feedback! Let me address your comment about page splits first. When a new page is added to the end of any level of an index (could be leaf or intermediate level), we do not move any rows around, we simply add a new empty page. The only time pages are left 50% full is when a new...
0 Likes
@Rmwin - good suggestions. For a sequential index yes the lowest level pages should not be 50% fill that's usually inefficient, except possibly for some odd configurations. However I believe the B-tree will still have to be re-balanced occasionally even with an optimized insert that minimizes half f...
0 Likes
Oh cool my comment lost all my formatting. I also seem unable to edit it... So a little correction. Where I wrote “Not many people what this issue”. I meant to write “Not many people have this issue”. Now I’ll see if more carriage returns allow me to format my text.
0 Likes