Hi Thomas ..thanks for your reply.

Immediate requirements ..option is to stah with VFP..mediumm term go Server route


We run vfp tables up into the 3.5GB area (dbf+fpt), but that is on single user access only (data transfer from big iron, some massage, basic data pruning and descriptive stats).


Have yiou a VFP .dbf table > 2 Gigs ? If so,this would solve the immediate problem .
Regards,
Gerard



>>Hi.
>>I have seen reference somewhwre recemtly that there is a way (and its not SQL Server !!) of maintining a VFP table greater then 2 gigs.
>>Anybody know anything about this or used it successfully ?
>
>Depends on your table - if you have many char fields not used in every query (they are markedly slower) you can switch those over to memo. Usually you get between 1.5 and 2 times reduction in size through the rtrim character of memo fields. Another option is to use 2 tables in a 1-to-1 relation; this is especially useful if your typical access pattern is through view or cursoradapter: you only have to rewrite that tiny part and the rewrite is straightforward.
>
>For more information worker habits it is usually safer to partition horizontally: packing only the data of one year or one country into the table and writing some intelligent acces code works.
>
>That said: If the data is important/expensive to recreate, consider using a C/S backend. DB2 entry version is free as well, supports nearly unlimited table space and IMHO reasonable restrictions on core and RAM usage. If you run into perf problems stemming from that limitation and cannot afford a non-free DB2 solution (or comparable MS-SQL), Postgres stays free<g>. If you are only considering in-house usage one MS-SQL server should pay for itself by saving work hours. I am not arguing that MS-SQL is "better", just that you will encounter not more and probably less problems taking dev time compared to the other solutions when accessing from vfp.
>
>We run vfp tables up into the 3.5GB area (dbf+fpt), but that is on single user access only (data transfer from big iron, some massage, basic data pruning and descriptive stats).
>
>my 0.02 EUR
>
>thomas