Thank you all for the input.
To answer all: I will be working on the product, Connect:Direct.
I once was a developer for it before IBM bought Sterling Commerce
from AT&T.
I had brought up the MVCL macro idea back then and got told
forget it.
I'm not looking to make a big splash, but I am checking to see
the experiences all of you have had in working on a performance
boost that may be possible.
There is an old adage: Don't fix what ain't broke.
Well, given the amount of I/O the product does, it is sensitive
to various changes that IBM has done. Splitting data from
Instruction areas (I-Bank/D-Bank) is really needed from my
experience working on old ALC code written from Base S/370
through ESA. Amazing what speed increases you get when you stop
having contaminated cache lines. Think in terms of OPEN in line
with other code that is NOT using a work area. The same for
PUT/GET, etc.
I hope most of what we have discussed here has already been
effected.
But like I said, wanted to get ideas from those of you that have
been dealing with these issues.
--
Regards,
Steve Thompson
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN