RE: RPG SQL and Cursors -- RPG400-L

The problem with optimization is: concentrate on those parts, which are
responsable for the runtime of the application!!!!

The obvious caveat being: Once the things you concentrated on have been
improved and optimized, the things you overlooked in order to concentrate on
those things will be the new bottlenecks. :) Iterative processes have
their place and that place is not program / system development.

Performance tuning is holistic, and every piece is important or none of it
is. You cannot get around it.

Dennis Lovelady
http://www.linkedin.com/in/dennislovelady
--
The 50-50-90 rule: Anytime you have a 50-50 chance of getting something
right, there's a 90% probability you'll get it wrong.

-----Original Message-----
From: rpg400-l-bounces@xxxxxxxxxxxx [mailto:rpg400-l-
bounces@xxxxxxxxxxxx] On Behalf Of dieter.bender@xxxxxxxxxxxx
Sent: Friday, November 19, 2010 2:11 PM
To: RPG programming on the IBM i / System i
Subject: Re: RPG SQL and Cursors

reading the post again, whats wrong with:

Up to now, I've ever succeeded in optimizing
Job runtimes ...

My native language is german, but I know the diffrence between ever and
never.

The problem with optimization is: concentrate on those parts, which are
responsable for the runtime of the application!!!!

Dieter
--------------------------------------------------
From: "Rory Hewitt" <rory.hewitt@xxxxxxxxx>
Sent: Friday, November 19, 2010 6:38 PM
To: "RPG programming on the IBM i / System i" <rpg400-l@xxxxxxxxxxxx>
Subject: Re: RPG SQL and Cursors

Dieter,

What do you mean, you've never succeeded in optimizing job runtimes?

Do

you
mean you've never made a code change which speeds things up? I *know*
that's
not true, because I've seen some of your code, and it's good stuff,

and I

can't believe it got that way without you figuring out how to make it
good...

I write a lot of code which, like Dennis's code, is used in

applications

where fractions of a second count. I've done a *lot* of work

improving

performance, and it's paid off handsomely.

Rory

On Fri, Nov 19, 2010 at 9:06 AM, <dieter.bender@xxxxxxxxxxxx> wrote:

... it doesn't help to get microseconds at one end, throwing away

seconds

or
even minutes at another end. Up to now, I've ever succeeded in

optimizing

Job runtimes ...

D*B

--------------------------------------------------
From: "Dennis Lovelady" <iseries@xxxxxxxxxxxx>
Sent: Friday, November 19, 2010 5:14 PM
To: "'RPG programming on the IBM i / System i'" <rpg400-

l@xxxxxxxxxxxx>

Subject: RE: RPG SQL and Cursors

That's cute and I look forward to writing such a note a

few/several

years
from now.

It's all relative. Microseconds may not matter in your business;
heaven
knows most System i are not exposed to high transaction volumes

per

unit
of
time.

If, however, you design your applications to be used by several
thousand
users concurrently, those microseconds shall add up. It will

matter.

It
does matter. Of course you have the option of never applying for

a job

at

a
place that does that sort of business.

Dennis Lovelady
http://www.linkedin.com/in/dennislovelady
--
Change is inevitable, except from a vending machine.

A little off-topic...

Some years ago Tracy Kidder wrote a wonderful book, "The Soul of

a New

Machine", regarding the efforts of Data General's computer

engineers

in
building of their first minicomputers. In that chronicle, of of

the

engineers charged with writing the computer's microcode got tired

of

working
with oscilloscopes and nanoseconds and, I quote,

*"He went away from the basement and left this note on his

terminal:

"I'm
going to a commune in Vermont and will deal with no unit of time
shorter
than a season." "*

Dieter, believe me, I really understand and share your

sentiments...

Best Regards,

Luis Rodriguez
IBM Certified Systems Expert - eServer i5 iSeries
--

On Fri, Nov 19, 2010 at 10:43 AM, <dieter.bender@xxxxxxxxxxxx>

wrote:

... I wouldn't use a mix of RLA and SQL, but maybe thats a

question

of

flavour and I'm tired of discussions about microseconds.

Dieter

--------------------------------------------------
From: "Aaron Bartell" <aaronbartell@xxxxxxxxx>
Sent: Friday, November 19, 2010 3:40 PM
To: "RPG programming on the IBM i / System i"
<rpg400-l@xxxxxxxxxxxx>
Subject: Re: RPG SQL and Cursors

I use a balance of RLA and SQL. Usually use RLA by default

and

only go

to

SQL when it is a better fit (i.e. dynamic sorts, searches,

ranges,

etc).

We
are actively moving our web development (i.e. OpenRPGUI.com)

from

our

V5R3

server to our V6R1 server so I am planning on jumping into

free

form

embedded SQL, and on that note...

Are there any well documented (i.e. articles) out there that

show

how a

CRUD
application would use embedded SQL for all file access? I

ask not

because

I
am new to SQL, but because I am not yet convinced of how it's
usage

is

being
described by some in this thread.

I must say that Alan's comments about lessening the compile

burden

is

something I hadn't considered to a great extent until I

started

thinking

about it. Certainly better that wrapping file access with

*SRVPGM's

where

the whole record is being passed in/out without applying any

sort

of

business logic. I think this also comes down to having a

good

change

management solution that does a lot of the work for you (as

it

should).

Aaron Bartell
www.MowYourLawn.com/blog
www.OpenRPGUI.com
www.SoftwareSavesLives.com
--
This is the RPG programming on the IBM i / System i (RPG400-L)

mailing

list
To post a message email: RPG400-L@xxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: http://lists.midrange.com/mailman/listinfo/rpg400-l
or email: RPG400-L-request@xxxxxxxxxxxx
Before posting, please take a moment to review the archives
at http://archive.midrange.com/rpg400-l.

--
This is the RPG programming on the IBM i / System i (RPG400-L)

mailing list

To post a message email: RPG400-L@xxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: http://lists.midrange.com/mailman/listinfo/rpg400-l
or email: RPG400-L-request@xxxxxxxxxxxx
Before posting, please take a moment to review the archives
at http://archive.midrange.com/rpg400-l.

--
This is the RPG programming on the IBM i / System i (RPG400-L)

mailing

list
To post a message email: RPG400-L@xxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: http://lists.midrange.com/mailman/listinfo/rpg400-l
or email: RPG400-L-request@xxxxxxxxxxxx
Before posting, please take a moment to review the archives
at http://archive.midrange.com/rpg400-l.

--
This is the RPG programming on the IBM i / System i (RPG400-L)

mailing

list
To post a message email: RPG400-L@xxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: http://lists.midrange.com/mailman/listinfo/rpg400-l
or email: RPG400-L-request@xxxxxxxxxxxx
Before posting, please take a moment to review the archives
at http://archive.midrange.com/rpg400-l.

--
This is the RPG programming on the IBM i / System i (RPG400-L)

mailing

list
To post a message email: RPG400-L@xxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: http://lists.midrange.com/mailman/listinfo/rpg400-l
or email: RPG400-L-request@xxxxxxxxxxxx
Before posting, please take a moment to review the archives
at http://archive.midrange.com/rpg400-l.

--

Rory Hewitt

http://www.linkedin.com/in/roryhewitt
--
This is the RPG programming on the IBM i / System i (RPG400-L)

mailing

list
To post a message email: RPG400-L@xxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: http://lists.midrange.com/mailman/listinfo/rpg400-l
or email: RPG400-L-request@xxxxxxxxxxxx
Before posting, please take a moment to review the archives
at http://archive.midrange.com/rpg400-l.

--
This is the RPG programming on the IBM i / System i (RPG400-L) mailing
list
To post a message email: RPG400-L@xxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: http://lists.midrange.com/mailman/listinfo/rpg400-l
or email: RPG400-L-request@xxxxxxxxxxxx
Before posting, please take a moment to review the archives
at http://archive.midrange.com/rpg400-l.