self related calcs

[culearn]

I am experimenting with calcs from self relationships, specifically for display in related files. I like the results but am unsure about a speed hit. Does anyone know if this is going to affect the speed?

Thanks

[LiveOak]

It's kind of a game of numbers. Let's use as an example an inventory system in which available quantity is calculated on the fly by adding all the + and - inventory transactions (instead of keeping a single number). This is a good general approach (transactions,that is) which allows entry errors to be discovered at a later date. If the inventory quantity available is just altered, you don't know how you got there.

The downside of this approach is that all these transactions must be processed to display an inventory quantity for an item. Looking at a single item, this might be ok for limited numbers of transactions. If you are either looking at items with a lot of transactions or a lot of items, this can drag to a crawl in displaying a list view. In most inventory systems this is too slow. A middle ground is usually to update a "working" quantity each night. This working number is debited or credited for real time transactions and updated by doing a complete processing of transactions each night.

Bottom line: in smaller solutions (records, number of related calculate field) this may be ok. In larger application it can easily be too slow. -bd