Radoslaw I agree that you wouldn't do this in production, but it is a perfectly valid way to measure the throughput of a host channel.
Besides fanning out through an 8Gb switch to multiple FICON blades or storage controllers, I would also suggest that you make sure the other port on the 8S channel card is not being used so that any shared paths and components do not affect your results. I always try to use only one CHPID per host card when I test in the HDS lab to minimize this sort of effect on measurement. Ron > > BTW: I'm going to connect the chpid to 8gbps switch and SFP and rerun the > test. Of course single channel attached CU is not intended to use in > production. > ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
