On Tue, 14 May 2019 09:35:42 -0600, Grant Taylor wrote: >On 5/14/19 7:08 AM, Tom Marchant wrote: >> Mildly? > >Yes, "mildly" is the word that I wanted to use. I explained why I chose it. > >> You can leave out the parenthetical "significantly". z machines can >> take a hard failure of a CP and a spare is switched in dynamically >> to take over the work. The unit of work that was running on that >> processor is moved to the new processor without interruption. There is >> a brief pause while it is switched, but the workload is not impacted. >> The operating system does not have to do anything to make this happen. >> It is done entirely in hardware. The failed processor can even be >> running critical operating system functions. It makes no difference. > >Said "brief pause" qualifies as a non-significant impact to me. Hence >the workload was impacted while it was moved from one CP to another.
Ummm... That's an interesting way to spin it. How long does it take for the hardware to reassign the currently running program to another CP? Microseconds? Nanoseconds? I don't know. The failing instruction is run on the new processor, followed by all subsequent instructions, until that program is interrupted. > >> Sure, detecting a potential failure situation and responding to that >> should be relatively trivial. >> >> That's the big difference, isn't it? > >Yes, it is a difference. What you said did not indicate to me that the >CPU had faulted. I guess you didn't notice "a hard failure of a CP"? >Rather I took what you said to mean that the CPU had a >cooling problem. Cooling problem? AFAIR, you are the only one who mentioned cooling problems. Nothing that I wrote was remotely related to a cooling problem. zArchitecture systems, and many generations of processor before that, responded to cooling problems in other ways. >That does not mean that the CPU is failed to me. Is >it usable as is? No. Is it spewing errors, shorting something, >otherwise adversely impacting the rest of the system? No. Wow. You have a strange notion of what a machine check means. -- Tom Marchant ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
