Application Module - Default Settings

[Saif Kamaal]

All,

Is it possible to benchmark the ideal configuration setting for the AM to say for about 100 concurrent users.
What i am looking out for is say the maxPoolSize, RerourcePoolSize etc. AM are heavy objects and proper configuration will really be beneficial and can be taken as a starting point for all projects.
Of lately a nos of post are floated on OTN forum regarding the proper values of these attributes and novice users are clueless of the same.

thnks
Kamaal

[John Stegeman]

Sure it's possible for (you) to benchmark it. Obviously, the "best" settings depend upon the application and the usage patterns, so saying "100 concurrent users" isn't really enough to make any useful determination.

John

[Chris Muir]

John, in order to give some guidance to people asking these sort of questions, what details should they supply?

CM.

On Thursday, 9 February 2012, John Stegeman <John.S...@gmail.com> wrote:
> Sure it's possible for (you) to benchmark it. Obviously, the "best" settings depend upon the application and the usage patterns, so saying "100 concurrent users" isn't really enough to make any useful determination.
>
> John
>

> --
> You received this message because you are subscribed to the ADF Enterprise Methodology Group (http://groups.google.com/group/adf-methodology). To unsubscribe send email to adf-methodolo...@googlegroups.com
>
> All content to the ADF EMG lies under the Creative Commons Attribution 3.0 Unported License (http://creativecommons.org/licenses/by/3.0/).  Any content sourced must be attributed back to the ADF EMG with a link to the Google Group (http://groups.google.com/group/adf-methodology).
>

[Saif Kamaal]

All,

What i really fell is if there is some best used configuration that experts have been using can really help a new developer.
For eg:
On the hardware front lets say we can streamline some settings for say 100 users with 1 weblogic sever on a 32 bit machine (so the heap size will be max of 1 gb) or a 64 bit machine (heap size of max 4 gb).
On the project level lets say 10 bounded TF with 10 AM (i have been brought up with the learning that 1 AM for 1 TF) etc, close to 10 fragments etc.

I agree that the values changes from application to application but i really feel there has to be a matrix of some of these values that give the best performance.

thnks
Kamaal

[John Flack]

My approach for pretty much anything that might need performance
tuning has always been:
1. Make some educated guesses about what good tuning parameters
should be, and deploy to test with those settings. If I'm not
educated enough to make a good guess, then I try the default settings.
2. Test, test, test. We've learned to use a good load testing tool,
like JMeter for ADF applications (thanks, CM, for your good advice).
3. If we notice performance problems in testing, we start diagnosing
where the bottleneck may be. Instrumentation (logging) is your
friend. Look for the same thing happening dozens of times, rather
than the long running process that only happens once, say at startup.
Don't immediately start guessing what will help - know where the
problem is first.
4. Often the bottleneck is in your code, not in the performance
parameters.
5. If it is in performance parameters, then change one and only one
parameter and retest. If you change more than one parameter, you
don't know what helped or hurt, and you bring into play interactions
that could complicate things. Go to step 2.
6. Tests look good? Then deploy to production, but continue to
monitor. Return to step 3.
7. Keep in mind performance goals - we're looking for "good enough" -
you need to know what "good enough" is, and when to stop tweaking it.

[Shay]

The best way would be to test your application.
We have the basic guidelines here:
http://www.oracle.com/technetwork/developer-tools/adf/learnmore/adfloadstresstesting-354067.pdf
And if you are interested Oracle also has a tool for this:
http://www.oracle.com/technetwork/oem/app-test/ds-ats-adf-testing-accelerator-200156.pdf