>Although, in almost every situation, SQL commands will be faster than going through a do while loop....
>
>>Ron,
>>*EACH* local machine executes its copy of the program on its local processor/RAM.
>>
>>In fact, each of the running programs is actually quite independent of each OTHER, relying on LOCK mechanisms to prevent problems.
>>
>>Your idea to "SET FILTER. . ." will not, in my humble opinion, help ANYTHING to do with this - each record will still have to be 'inspected' by the machine issuing the SET FILTER to see if it "fits" or not, so *ALL* data still has to come down the pipe and into the local machine.
>>
>>Hope this helps,
>>
>>Jim N
>>
>>>Dear all,
>>>
>>>This question has been bothering me for a while.
>>>Considering the architecture, more than two computers, running windows 95, are connected together by LAN. A program and associated database are put on one of the machine, let's say 'machine A'. Other machines remotely executes the program on machine A through network neighbour, and shows the forms and results on their local screen. What happen to this? Is the program actually exectued on remote machine A, or run on local machines? Or some things are run remotely, and some other things are run locally?
>>>
>>>I have an experiment about it, and it shows the resources on local machines are used quite a lot. But I am not sure why.
>>>
>>>Also, in this case, is it better to issue SQL-commands than using 'set filter to & do while loop' to manipulate data? Considering the performance and the amount of data to deal with.
>>>
>>>Many thanks.
>>>
>>>Regards,
>>>Ron.

Not true...SQL commands are not necessarily faster. It depends on;

1) The amount of data
2) Proper indexing
3) Amount of virtual memory on the PC
4) How the SQL statement is structured

At a previous job, I had an 11,000,000 record table. SQL came to
a CRAWL when using this table. SEEK and SCAN were *MUCH* faster.