@Robert Garrett (with a sigh of relief!) 15 minutes for z/CX to come up on a z/OS under z/VM at Dallas? Me too. I thought I was doing something wrong, but couldn't get any useful feedback from IBM/Dallas about the 'problem'.
Sean On Thu, 21 Apr 2022 at 04:57, kekronbekron < [email protected]> wrote: > Apologies if this seems rash. > Certainly don't mean to belittle people's work; many are restricted with > choices, procedures, etc. > If it isn't for the cost of being an MF s/w vendor, competent new > solutions would steal the show. > Much like most of y'all, I want Z to remain king of the hill. > > -KB > > ------- Original Message ------- > On Thursday, April 21st, 2022 at 9:06 AM, kekronbekron < > [email protected]> wrote: > > > > As an observer, I reckon IBM are forced to use OpenShift because they've > got to get RedHat in there. > > Also, since everyone knows the word Docker now, the Z "has to have it". > > Surely the industry is now waking up to the mess that is the Kubernetes > ecoystem mgmt., service MESS. > > > > I do wonder... for those who thought setting up zOSMF RACF was painful, > what their journey will be for zCX and OpenShift. > > Just saying, "it's free because zIIP" doesn't make it good. > > No Ferrari owner should be "compelled" to use a unicycle's wheel just > because it's free. > > > > IMHO, "Me too" solutions are seriously ruining the reputation of the Z > with the ridiculous CPU, memory, storage requirements. > > I thought it was ridiculous that RDz wanted a few gigabytes of memory > for the JVM. > > Rebadged oldware, with web stack & interface from early 2010s, are now > coming to compete with Chrome, in their lust for memory and such. > > > > - KB > > ------- Original Message ------- > > On Thursday, April 21st, 2022 at 4:22 AM, Tony Harminc [email protected] > wrote: > > > > > > > > > On Wed, 20 Apr 2022 at 18:00, Robert Garrett [email protected] > wrote: > > > > > > [...] > > > > > > > It likes a LOT of real memory and it appears that the running > instance consumes the full amount of real memory allocated to it for the > duration, making it unavailable to zOS for paging or any other use. Don't > believe the claim that it can run in as little > > > > as 2GB. The experimental test instance that I built had 3GB > allocated to it (the most I could give it on the LPAR I was using) and it > took a full 15 minutes (yes minutes) by the clock for the address space to > initialize and reach the point where it was > > > > functional - on every start up. Admittedly, this was on a zOS image > that was being hosted under zVM at IBM Dallas, so I'm sure that had some > impact. > > > > > > That smells like three levels of SIE, which to my understanding is > > > never going to perform reasonably. > > > > > > Tony H. > > > > > > ---------------------------------------------------------------------- > > > 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 > > ---------------------------------------------------------------------- > 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
