Where two similar fields exist in the Subset, does the choice of which field to use have any effect on performance?
In most cases, the answer is yes. The most common reason for apparently identical fields to exist in a Subset is where the field is a key in one or more files. For example, customer number might be a field in the order header, but it is also the key to the customer master file. Consequently, there are two fields defined in the Subset called customer number, and the choice of which field to use can be significant. Consider a report that lists some fields from the order header file. No customer details are required, only the customer number. If the customer number from the customer master is chosen instead of the customer number from the order header, then OPEN/DB thinks that the customer master file is required and will include it in the report – normally, it would be omitted. Note that this problem is not functional – whichever customer number is used, the resulting report will be correct. However, there may be an effect on relative performance. In most cases, this type of problem can
