The only "solid" research that I have found on fragmentation is dated 1997. In the end, anything that I've come across that's been written since then ultimately references that document (if it references anything).

As regards "solid" research on (VFP) REINDEX before/after defragmenting, I'd say there is none.

Here are some things to consider though:

1) The Disk Defragmenter utility that comes with Win2K and WinXP (a cut-down version of Diskkeeper, by all acounts, so you should expect similarly for that utility) does the job on a "best fit basis only. This means that there is a chance that your .CDX, .DBF and .FPT for any given file can be spread (each contiguously) all over the occupied space on the drive.

2) REINDEX (except in VFP7SP1) will remove CDX "bloat" of both the deleted TAG variety and the 'random inserts causing extra nodes' variety. A smaller file is always better and I should think especially so where it is a .CDX. A REINDEX will also result in a 'contiguous' (as much as available RAM will allow) .CDX file as its output.

good luck in your decision.


>Has any solid research been done in reference to performance after re-indexing before verses after defragmentation of the drive?