Please start any new threads on our new
site at https://forums.sqlteam.com. We've got lots of great SQL Server
experts to answer whatever question you can come up with.
| Author |
Topic |
|
rubs_65
Posting Yak Master
144 Posts |
 Posted - 2003-10-01 : 12:05:31
|
| Hi,We are planning to rebuild our indexes as they are too fragmented. I need some suggestion for setting of fillfactor for clustered and non-clustered indexes. We have 3 types of tables:1) Large tables having about 20 million rows and we add about 1-2 million rows per month. only select/insert are issued on these tables never update and delete. also insert are in sequential order as primary key is on identity column..so rows are always inserted at end...2) small tables having about 50000-100000 rows and we add about 5000-10000 rows every month; insert/select/update is issued on tables.mostly insert and select ; update sometime but not too often..rows are inserted at end as primary key on sequential column and never use the value again(if deleted)3) very small tables <2000 rows ; all dml operations performed.What should be the fillfactor values for these 3 different set of tables?Thanks--rubs |
|
|
samsekar
Constraint Violating Yak Guru
437 Posts |
Posted - 2003-10-06 : 07:49:13
|
| you have to take considerations on which column you are going to create indexes. how you are going to select data from your table. if you are going to have clustered indexes on identity columns and you might need to have higher fill factor, which gives high performance on SELECT. Check BOL on "fill factor" for more details.- Sekar |
 |
|
|
|
|
|
fillfactor suggestion