Performance Questions and a little need for some education. -Output Phase reading

[Peter Bateman]

 Hi All:

 

   If you really need to do large sorts you can improve your performance by 

   writing to a subfile and then doing a VMS sort over which you have more control.

  

   This issue has come up on this list before.

   Check out http://lists.sowder.com/pipermail/powerh-l/1998-April/000247.html

 

Regards,

Peter


From: ken.langendock at rogers.com
To: powerh-l at lists.sowder.com
Subject: RE: Performance Questions and a little need for some education. -Output Phase reading
Date: Wed, 19 Aug 2009 10:54:32 -0400







Ah sorry, that still counts as a file in the file counter…
 
Speed is the biggest improvement. Clean code is just a benefit of doing it this way.
 
BTW, I didn’t come up with it all by myself. I used to work at Cognos for a brief period of time in the Toronto office a loooooooonnnnnnggggg time ago and a brilliant programmer by the name of Doug Blackmore showed me how to do this. He was in my opinion one of the top 3 PH programmers in the world at that time. I think he is still there, but they moved him into pre-sales…what a waste.
 
Ken
 


From: Dan Lambshead [mailto:dlambshe at milk.org] 
Sent: August 18, 2009 4:36 PM
To: Ken at Langendock.com; powerh-l at lists.sowder.com
Subject: [Bulk] RE: Performance Questions and a little need for some education. -Output Phase reading
 
Hi Ken,
 
  Yes, that is exactly what I've seen! I would say it makes the code more readable, by simplifying the Access statement, and it also reduces the record complex so could improve performance. Beyond that I am trying to remember if there are any more advantages (perhaps for one, if you're reaching the no of files limit, it doesn't count).
 
Dan
 



From: powerh-l-bounces+dlambshe=milk.org at lists.sowder.com [mailto:powerh-l-bounces+dlambshe=milk.org at lists.sowder.com] On Behalf Of Ken Langendock
Sent: Tuesday, August 18, 2009 4:20 PM
To: powerh-l at lists.sowder.com
Subject: RE: Performance Questions and a little need for some education. -Output Phase reading
Yep that is what I meant.
 
The advantages:
If you relate this to a QUICK screen; you have a Primary file and some detail/secondary files and reference files.
 
In QTP the only thing you need to access (via the ACCESS statement are the Primary and/or Secondary/detail files.
All other files can be read in the OUTPUT PHASE (obviously that rule is subjective depending on the desired record complex).
 
For example:
In a screen you could have:
FILE Orders PRIMARY
FILE OrderDetails DETAIL
FILE Products REFERENCE
Show the order info, detail info and the product description…etc.
 
 
In QTP you can set the same thing up like this:
ACCESS Orders &
  LINK OrderNumber OF Orders &
    TO OrderNumber &
    IN OrderDetails
 
  SORT ON ProductNumber OF OrderDetails &
       ON DateLastOrder OF OrderDetails
 
OUTPUT Products &
       AT ProductNumber &
       ADD &
       UPDATE &
       NOITEMS &
       IF 1 EQ 2 &
       VIA   ProductNumber &
       USING ProductNumber OF OrderDetails
    
 
SUBFILE subfile1 &
        KEEP &
        AT ProductNumber &
        INCLUDE whatever
 
 
Ken
 


From: Dan Lambshead [mailto:dlambshe at milk.org] 
Sent: August 18, 2009 3:25 PM
To: Ken at Langendock.com; powerh-l at lists.sowder.com
Subject: [Bulk] RE: Performance Questions and a little need for some education.
 
Hi Ken,
 
  Just wondering what you mean by "Output Phase Read". Is that when you do an Output ... if 1 = 2 ... or something like that, so you are only doing a read? I have seen that before, but now am trying to remember what the advantages are (if that's what you're talking about).
 
Dan
 
_________________________________________________________________
Stay in the loop and chat with friends, right from your inbox!
http://go.microsoft.com/?linkid=9671354
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.sowder.com/pipermail/powerh-l/attachments/20090819/780db1b6/attachment-0001.htm