Interesting article, though I would have appreciated a bit more information to see the while picture, most importantly the external fragmentation of indexes, but also the internal fragmentation. Information about he growth settings for the database files, and whether it has been shrunk (or even autoshrunk) would be nice too.
Now, the rest of this post is meant to give those interested a bit of extra resources, and possibly explain why the fragmentation (both of indexes and file system) could cause issues.
I may be wrong, but I would expect your issues to be caused by a high external fragmentation[/url] causing a lot of seek back and forth in the database file, possibly combined your disk fragmentation causing a lot of entries in the NTFS MFT[/url] (Master File Table) and most importantly causing a loot of seek on the disk system. Other factors possibly affecting your performance would be incorrect partiton aligmnent, and the default 4K allocation unit (instead of 64k), both of which could cause a single logical read to become multiple physical reads.
Well, I hope this information is useful for someone 🙂
Ole Kristian Velstadbråten Bangås - Virinco
Concatenating Row Values in Transact-SQL[/url]