Re: RUNPTY and TIMESLICE in *CLS objects

Hello Nathan,

Am 16.12.2021 um 19:31 schrieb Nathan Andelin <nandelin@xxxxxxxxx>:

I used to run a PC based stress-test tool to test the performance of my IBM i web applications, which handled HTTP POST and GET methods. The tool could drive IBM i CPU utilization to 100%.

Hint: There is also a command line "ab" tool, Apache Benchmark available.

https://httpd.apache.org/docs/2.4/programs/ab.html

It was interesting that smaller time slices for those IBM i workloads resulted in the highest overall throughput. I eventually reduced the time slice to 1 millisecond, which provided the highest throughput.

This highly depends on the machine you have used. Maybe you can tell us a bit more about it?

That seemed counterintuitive to me.

In a way, it is.

Frank Soltis asserted that IBM i provided faster task-swapping than any other operating system out there.

Because there is no swapping (given there is enough main storage available to keep everything needed accessible without paging activity). From what I've understood from his "Fortress Rochester" book, a context-switch is merely the rewriting of the address from where the CPU should fetch it's next instruction. I'm almost 100% sure, this is grossly oversimplified, though. I'm not even sure if the concept of a context switch is applicable to IBM i and predecessors.

https://en.wikipedia.org/wiki/Context_switch

:wq! PoC