• Subject: Re: Error Indicator on Write
  • From: "David Morris" <dmorris@xxxxxxxxxxxxx>
  • Date: Tue, 16 Mar 1999 09:35:28 -0700


Please don't take my comments as any kind of criticism, that would be like 
biting the hand that feeds 
you.  I saw the SETLL/READ sans (E) and was trying to rib you because of the 
"definitive" statement.  
I used to worry about whether one opcode was faster than another and have come 
to believe that it 
doesn't matter much, because I/O almost always sets the pace.  About the only 
optimization that seems 
to help is to avoid moving or comparing large amounts of data.

The I/O routines I wrote sit between an application and database server.  These 
routines support things 
like GetRcd *NEXT, *FIRST, *EQUAL... you get the idea.  I have found this extra 
layer adds a negligible 
amount of overhead, which is offset somewhat by the ability to really optimize 
I/O. Yes, I replaced the 
routine that used to trigger a READE which was a *NEXT w/key with a READ and 
key comparison.  I 
am now exploring using the C library I/O routines and dumping the generated RPG 
I/O modules.

David Morris

>>> "Simon Coulter" <shc@flybynight.com.au> 03/15/99 05:21PM >>>
Mello David,

I agree with you.  Using SETLL and READ(E) in place of CHAIN is just dumb -- 
although common enough.  Proper 
use of blocking is generally the best single performance improvement that can 
be made to an application.  It 
does rely on sequential processing and may need a sort of some sort.

My original comments were specifically regarding the proper way to handle 
WRITEs and whether duplicate record 
checking should be performed before the WRITE.

Regarding your I/O routines:  Are you replacing an incoming request for 
read-equal-key with your own 
sequential table scan and a large blocking factor?

Simon Coulter.

 FlyByNight Software         AS/400 Technical Specialists       
 Eclipse the competition - run your business on an IBM AS/400.  
 Phone: +61 3 9419 0175      Mobile: +61 0411 091 400           
 Fax:   +61 3 9419 0175      mailto: shc@flybynight.com.au      
 Windoze should not be open at Warp speed.                      

| This is the Midrange System Mailing List!
| To submit a new message, send your mail to MIDRANGE-L@midrange.com.
| To subscribe to this list send email to MIDRANGE-L-SUB@midrange.com.
| To unsubscribe from this list send email to MIDRANGE-L-UNSUB@midrange.com.
| Questions should be directed to the list owner/operator: david@midrange.com

This thread ...

Follow On AppleNews
Return to Archive home page | Return to MIDRANGE.COM home page

This mailing list archive is Copyright 1997-2020 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].