The system I am converting to client server is a vehicle loading software package.

For each location, I receive a file containing the days loads which can consist of 5 to 50 routes as a rough example. Each route can have several stops. Each stop can have several customers and each customer can have several orders. Each order consists of one or more products.

It is not unusual for a single location to have a load file that has 10000 lines (a line representing one product). Most of the parameterization in this project is to seperate the individual locations from one another.

The general system consists of a number of setup files - products, packages, vehicles, employees, etc. Then there are the short term processing files - load request, route problems, etc. These short term files hold the data for at most a few days. For the most part it should only hold the data for one day as the loads will be built that night and shipped the next. But we do allow for them to sit for a few days unless the user archives or purges. The adding of records seems to be reasonable, but the deletion of records makes the tableupdate drag.

There must be something I am doind wrong with views.

>Jason,
>
>I can't say that I've ever tried to use such large updateable p-views. A p-view is supposed to be some subset of records from the real table, you can do updates/inserts/deletes and then commit the changes back to the table.
>
>Why is your table empty and your p-view 7k records long? Is this like an initial load of the table? Why not do the inserts straight to the table?
>
>Can you restructure your code to do smaller chunks of records, say 100 at a time does the performance radically improve?
>
>>Yes most of the fields are in the PK.
>
>Integer keys would make your life easier. What you have there looks more like a sort by clause used at reporting time.
>
>>At current I am going from a completely empty table.
>>
>>The table is on a local drive.
>>
>>The parameterized view has all the records.
>>
>>The two functions I am testing is importing new data and purging the data.