Concurrency in QSRT sitting on a OS/390. How is it handled?

[ferreirahe]

Hi guys,

I'm comfortable using QualityStage RealTime in Windows and Unix, but now I need to implement it on OS/390, which is quite different.
I've got no problems setting up QSRT Server and QSRT Manager's JCL.

The issue is to understand how it deals with multiple concurrent requests:

- How does the QSRT Manager behave when it gets multiple requests at the same time? Does it keep them on queue and handle them sequentially? Or does the client fails to connect?

- Does QSRT Manager and QSRT Server behave identically to Windows and Unix version? Are there the same tuning options (MAXPOOL, TRANSQUEUE, etc.)?

- If a client made a request to a given instance (by TAG), but that instance is already processing another request from another client, how it will be treated? Fork to process the request in parallel? Process the request sequentially? Or denies the connection to the new client?

- If the behaviour is similar to Windows/Unix version, it'll fork to deal the request in parallel. If so, which TCP/IP port will be used? All TCP/IP ports that may be used need to be previously request to the System Administration team.

I wonder if one of our dear "gurus" can help me out here. I am really lost.

Thanks in advance & best regards,
Henrique