On Saturday, 10/24/2020 at 01:03 GMT, "van Sleeuwen, Berry" <[email protected]> wrote: > Hi All, > > The PIPE VMC stage apparently has a fixed 8K buffer, though I don’t know for > sure if that limit is in the VMC stage or a limit in the Performance toolkit. > When retrieving data it only returns 101 records of 80 bytes as 101 records > fills that 8K buffer. > > When I run PIPE VMC USER with more than 100 users in z/VM it will discard the > last records. And in our CPC we have a number of LPARs, with multiple CPUs, so > PIPE VMC LPAR also discards the records after record 101. > > Is there a way to increase the buffer size, perhaps similar to the buffer size > of the CP stage?
It's hard-coded at 8K. It's weird that PIPE VMC is documented in the PerfKit books as being an interface, yet also says, "The data transfer buffer that is used by CMS Pipelines has a size of 8kB, allowing transfer of up to 102 lines of 80 characters." What the point is of an interface that can't return all of the data? It seems to me that the VMCX command included with PerfKit should include an option to return all of the data without highlighting, etc.. Alan Altmark Senior Managing z/VM and Linux Consultant IBM Systems Lab Services IBM Z Delivery Practice ibm.com/systems/services/labservices office: 607.429.3323 mobile; 607.321.7556 [email protected] IBM Endicott
