True, but they live on the DASD. Depending on the install method the vendor uses, the dataset can stay a short time or a long time. Plus when using the RECEIVEd datasets, poorly blocked datasets, again with large datasets, you do more I/O and like I keep saying, CPU drives I/O. I have run the vendor's poorly blocked datasets vs optimally blocked datasets and the EXCP and CPU (TCB and SRB) savings are noticable.
"Confidentially doc, I am the wabbit." Bugs Bunny Sent from Proton Mail for Android. -------- Original Message -------- On Thursday, 12/18/25 at 17:10 Jon Perryman <[email protected]> wrote: On Thu, 18 Dec 2025 21:16:22 +0000, rpinion865 <[email protected]> wrote: >Broken record here. But I see our QR libraries with the following DCB >information FB 80/6160 >non-loadlib datasets and U 0/6133 for load libraries. The libraries are not >large, but still >why not join at least the 1990s. No one seems bothered with this limitation. Transmit files are sent using the internet where packet sizes are smaller than this size. These files are being used as a transmission medium instead of long term storage. ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
