>Rod,
>
>The performance hit with clustered index, and it's a bad one, occurs whenever a save operation occurs on the document table. A document save is typically for several dozen records, potentially hunderds of records, at a time.

Steve,
In the case of batch updates, have you attempted dropping the indexes before the batch starts and rebuilding when complete? If you insert a row that needs to be between two other rows in the clustered index (CI), then SS modifies the CI as well as any NCIs that refer it. In your case, this could occur dozens or hundreds of times. On a table with 800,000 rows, this would be a huge performance hit. If you drop the CI and then rebuild it, all the processing would take place once before to modify the NCIs to refer to the RID, as Travis posted, and once after to make them point to the CI.