I have very little data that was dependent on arrival sequence. Normally
limited to EDI exchange files.
As far as crossing my fingers and hoping that a select without an order by
would bring it up in arrival sequence, I've never found having to use the
order by clause to be an inconvenience. Nor hoping that a DSPPFM would be
orderly. On those that I had some mild care that the DSPPFM would bring
it up in some order I used the KEYFILE parameter on RGZPFM. Using the
KEYFILE parameter may increase performance because you can get better
blocking if the file is sorted by a popular access path. For example if
you are processing a table where the active record code is '1' and the
order of the item number then using an index by active record code and
item number as the KEYFILE parameter in RGZPFM should assist in that
I would think that without reusing deleted records you may have a
performance hit due to the need to create extents as rows are added.