January 22, 2002 at 4:10 pm
I have to standardize the index names on the development database and found it difficult to create a naming standard due to the multiple columns for each of the indexes. We have standardized on the Unique Clustered
Index for our primary key implementation and the naming standard for this table is pretty straight forward.
For example : For a table named Company_Group we have the Clustered Unique Index as PK_Company_Group and for Cash_Request_For_Billing we have the Clustered Unique Index as PK_Cash_Request_For_Billing.
The problem is for the other indexes that needs to be created for optimization processes. While testing and optimizing, I created indexes with random names such as x, t ... etc since the indexes were based on multilple columns. Also the Index Wizard creates names such as tablename6, tablename66, tablename666 ... etc.
I have thought of naming indexes with three characters for each column. So if the index on the table Employee is based on the Employee_ID, Employee_Age, Employee_Phone then I would name the index as NUI_EidEagEph
-where NUI stands for Non-Unique Index. But the names tend to become awkard as the number of columns in the index increases to five or six.
It would really help if others could share their naming standards.
THanks.
January 22, 2002 at 4:13 pm
Since I don't work with indexes constantly, I just use _NDX1, NDX2, etc. For clustered I use _IDX. Holdovers from the past.
I tend to tune and then let things alone so having a naming convention to decode isn't worth it.
Steve Jones
January 22, 2002 at 5:12 pm
I tend to use ndx_somethingdescriptive, usually col names, or maybe 'nodupes' if specifically to prevent duplicate entries (not on a key, but some other value). Dont really care though!
Andy
Viewing 3 posts - 1 through 2 (of 2 total)
You must be logged in to reply to this topic. Login to reply