>Testing is necessary part of development process, but it's impossible to simulate real situation 100%, because of many reasons: network traffic obviously will be different, number of deleted records may vary significantly during the system life span etc.

I think it is possible to test if a DELETED() tag would benefit you're performance. And if you don't find any major difference then why bother using the index ?

>In regard to Rushmore: it seems that nobody knows exactly how it works, but assuming that some real index files are tens MB on network, one can say that moving whole index to local computer (hard drive in this case) and only then starting processing is not the way Foxpro actually works.

I think, VFP loads the only the index records by which the query could be optimized. IOW, if the query is about invoice# 1000, it loads the indexrecords regarding invoice# 1000. If the query is about invoice#>1000 and clientnr=5000
it loads indexrecords regarding invoice#>1000 and indexrecord regarding debnr=5000. The following step is to combine those two types if indexes (in a index query) and use the result to gather the actual resultset.

But when there is not enough memory to load all the appropiate index records VFP is not able to use rushmore to optimize.

When using an index on deleted on a large table all the indexrecords regarding to non-deleted records should be in memory. Because of that in this type of index you can only have .T. or .F. this would be the majority of all the indexrecords of the deleted tag(). This could theoreticly lead to the disability of rushmore to optimize.

This is only my own theory; maybe someone could confirm or deny.


walter,