Ok, I see now. Yes, we should probably have put a caveat on that, and I think the confusion is partially because everything Bill says in the presentation isn't on the chart. There are two separate factors:
We have indeed run experiments with sizes greater than 512GB internally, but those experiments were run with an unpublished modification to the initialization code, removing the cap enforcement, just to "see what would happen". Also, that portion of the chart is referring to the maximum LPAR size on the particular machines, which is an entirely different question than where the size capping is enforced, but since they're both in "unsupported territory", it would perhaps be confusing and too much detail to cram all on the existing chart. Bill tells me this is an older version of the chart. He will try to get the most current one out there today, and correct some of the concerns raised here. - Bill Holder, z/VM Development, IBM Endicott On Mon, 11 Oct 2010 15:07:29 -0600, Scott Rohling <scott.rohl...@gmail.co m> wrote: >I got them from Bill B's 'VM Limits' presentation .. 512GB for z9 -- >1TB for z10 .... > >http://www.vm.ibm.com/devpages/bitner/presentations/vmlimits.pdf > >Scott Rohling >