Re: Java on 7.1 one system fast other system very slow

[Vernon Hamberg <vhamberg@xxxxxxxxxxxxxxx>]

Hi John

I suggest you run database monitors from Navigator over the jobs in 
question, on each system - then do the Visual Explain thing - you'll be 
able to see what optimization choices were made, and maybe see a 
difference there.

Good luck, and come on back with the results, then we can speak to this 
better, I think.

Vern

On 8/17/2015 4:01 PM, John Allen wrote:
> We have a CL program that calls a java program to read modify and
> write a file in the IFS
>
>   
>
> Both machines let's call them F and S (Fast and Slow)
>
>   
>
> They are configured the same (except slight difference in memory and
> disk storage see below)
>
> Both running IBM I 7.1
>
> Both at the same PTF levels for both the SS1 and JV1
>
> Both have similar memory
>
> Both have similar System ASP used (10 % vs 23%)
>
> Both have Java Version =
>
> java version "1.6.0"
>
>
> Java(TM) SE Runtime Environment (build pap3260sr16fp1-20140706_01(SR16
> FP1))
>
> IBM J9 VM (build 2.4, JRE 1.6.0 IBM J9 2.4 OS/400 ppc-32
> jvmap3260sr16fp1-20140706_01 (JIT enabled, AOT enabled)
>
> J9VM - 20140626_204542
>
>
> JIT  - r9_20130920_46510ifx7
>
>
> GC   - GA24_Java6_SR16_20140626_1848_B204542)
>
>
> JCL  - 20140704_01
>
>
>   
>
>
> Note:
>
> Executing JAVA *VERSION
>
> System F response is instant
>
> System S Response is 20 seconds
>
>   
>
> We are seeing something similar with our CL program
>
> System F= 5-10 seconds
>
> System S=60-80 seconds
>
>   
>
> The "JAVA *VERSION" execution is so different I don't think it is
> anything in our CL or java program.
>
>   
>
> I keep comparing the 2 and can't find a difference other than the
> performance of Java being so vastly different
>
>   
>
> What could cause a variance of the simple execution of "JAVA *VERSION"
>
>   
>
> Any idea or suggestions on where to I can look?
>
>   
>
> Thanks in advance for any assistance
>
>   
>
> John
>
>   
>
>   
>
>   
>
>   
>
>   
>
>