Mapping tool very poor performance with display queue fonctionality

[Former Member]

Hi all,

I've a very borring issue with debug tools offered by test mapping tool.

It's unworkable because when we want to see the displayed queue on any steps, it takes more than one minute !

PI version is 04s 7.0 SPS12.

We intend to patch it to SPS14 but we don't know if this will solve this.

Any idea ?

Thanks a lot.

J.C

[Former Member]

hi jean,

hope you are using diaply queues for lookups JDBC or RFC

if that is the case then you need to check with network between your XI and the servers.

due to network u may get delayed responses,we faced this problem.

if it is context mapping then you need to look at hardware enhancemet than patch level.

XI tuning ll require better RAM and CPU speed.

note:if you call thirdy party applications then display queue ll be longer in duration.

Thanks & Regards,

Rama Krishna

[Former Member]

Hi,

Could you plz explain in detail what you are trying to debug that you say is too slow?

Remember that Integration Builder is a java web start application, and most of its functionality runs locally. So, your desktop configuration can influence directly the final performance you have.

I worked with same version you have and never got this issue.

regards.

roberti

[Former Member]

Thanks for your response.

I just want to debug a message mapping using "display queue tool fonctionality".

Each time the designed mapping node is complex (except constant affectation or direct node affectation), results take a very long time to be displayed !

My local developper computer :

Pentium 1.73 Ghz, 1 Go RAM.

Thanks

[Former Member]

Hum...

could you please specify what do you mean with "complex" (post a little sample). if you are just using standard mapping tool functions it would not be slow... but if you are using your user defined functions, then you need to check the performance in that functions, probably there is the bottleneck.

when you are just testing, do you use large number of records? like, more than one thousand of records?

regards.

roberti