|
>While we're on the subject, which is better performance wise, having the >smallest file as the primary joined to a larger file, or visa versa? >I remember being told this many moons ago. Left File(1st one defined in Query/400) should be the smaller. Remember, That's the smallest after any where clauses. >And another bottom feeders question, performance wise, am I better off >querying a logical file than the primary? I expect yes, but ignorance is not >bliss. As a rule of Thumb, Chose the PF. The optimizer should find the one you need. Remember, the indexes are looked at chronologically(the last ones created are the first ones looked at) So if the most useful LF/Index is the first one you create, and you have large number of indexes (over ????? now comes the posts on "We have more than that") The optimizer may timeout before finding the "Perfect" choice. If you are running an SQLRPG program, do a PRTSQLINF and see some interesting history of past run times. Before running your program or before running even Query/400 do a STRDBG. Then Look at your job log. You will see messages like CPI432C which tells you why it didn't use certain indexes. Also if there is only (say) a few thousand records in the file, It doesn't matter what file (PF or LF) you point at, the optimizer will decide that it would be faster just to do a full data base scan(ie read each record) instead of wasting time bringing an index into memory, etc. etc. Hope this helps John
As an Amazon Associate we earn from qualifying purchases.
This mailing list archive is Copyright 1997-2025 by midrange.com and David Gibbs as a compilation work. Use of the archive is restricted to research of a business or technical nature. Any other uses are prohibited. Full details are available on our policy page. If you have questions about this, please contact [javascript protected email address].
Operating expenses for this site are earned using the Amazon Associate program and Google Adsense.