Slow processing of messages - APO to PI Single stack

[Former Member]

Dear Experts,

We have a requirements to send more than 10000 records as individual messages from APO to PI using outbound proxy.

The processing of message happens sequentially via dedicated queues in APO(SMQ2).

When outbound proxy job executes, thousand of messages piled up in SMQ2 in running status and schedule status in SXMB_MONI of APO due to which there is always a delay in interface execution and breach SLAs. But, in PI, there is no delay in processing of message upon reaching from APO.

Is there any issues with APO or PI performance ? Please provide your opinion from past experience and recommendation to improve the performance of the interface as it is a very critical to complete on time.

Regards

Danish

[engswee]

Hi Danish

You might need to tune your PI system accordingly. I'd suggest reading the following two articles to have a better idea of performance in PI and tuning.

In particular, for your case since you are on single stack, you would be using a SOAP sender channel with XI 3.0 protocol for the outbound proxy scenario. I'd suggest increasing

i) worker threads of the SOAP adapter - refer to "Configure consumer threads per adapter" in first link above

ii) FCA threads - refer to section 6.1.2.1 of document in second link above

Rgds

Eng Swee