SQL Server Performance Forum – Threads Archive

Index skip scan feature.

1. In Oracle 9.2+ you have:
"Index skip scans to improve index scans by nonprefix columns since it is often faster to scan index blocks than scanning table data blocks".
.
-> This means that if I create an index like this one:
CREATE NONCLUSTERED INDEX [IDX1] ON [dbo].[VT_NUTR](
[VTNST_NUTRG_ID] ASC,
[VTNST_NUTR_ID] ASC)
.
2. Then later I can create a where clause that only supplies the 2e part of the index, i.e. VTNST_NUTR_ID.
-> And still the index is used.
.
Does SQL Server 2005 (whatever version) has this feature as well?
(The alternative is to create two separate indexs. The advantage of 1 index is that you can makte it a primary key as well).
.
Regards,
Bert

---

]]>