| SQL Server 6.x |
SQL Server 2000 |
| The SORTED_DATA_REORG clause of CREATE INDEX eliminated the sort performed when a clustered index was created. |
Replaced by the DROP_EXISTING clause of CREATE INDEX. Remove all references to the SORTED_DATA_REORG clause of CREATE INDEX and replace with references to DROP_EXISTING. |
| The SORTED_DATA clause of CREATE INDEX eliminated the sort performed when a clustered index was created and physically reorganized the data. |
Removed; no longer available. Remove all references to the SORTED_DATA clause of CREATE INDEX. |
| bcp could import an already sorted data file into a SQL Server table. Creating a clustered index on an ordered table could be optimized by using the SORTED_DATA clause of CREATE INDEX. The SORTED_DATA clause forced SQL Server not to sort or reorganize the previously ordered table. |
SQL Server returns an error message stating that the SORTED_DATA clause of CREATE INDEX is ignored and no longer supported. Remove all references to the SORTED_DATA clause of CREATE INDEX. Consider creating the clustered index before using bcp to import the data. bcp uses improved index maintenance strategies to make data importation with a preexisting index faster than earlier releases and avoids resorting of data after importation. |
| The ALLOW_DUP_ROW and IGNORE_DUP_ROW clauses of the CREATE INDEX statement allowed data to be updated into tables with a unique index and without having to filter out duplicates first. |
No longer supported. Using either ALLOW_DUP_ROW or IGNORE_DUP_ROW in the CREATE INDEX statement generates a warning message. If there is no unique clustered index and there is a need to avoid duplicate rows, create a unique constraint on one or more columns other than the clustering key. |