IBM 610 workstation computer 3432
It is a poor man's fault tolerance. SMP makes a system able to withstand faults.
Why do you think I'm fighting with all you guys? This hot-plugging should have happened 2 decades ago. TOPS-10 delivered a way to do something similar to a hotplug; the restriction was that there would be a momentary pause in delivering computing services, but only in certain cases. IIRC, the only reason to actually pause the monitor was to arrange memory. Everything could be done gracefully with at a CTY command level.
PTUI! This was...sort of...in 1980 SMP. I was beginning to work on TW to change his atbreastude about allowing dynamic additions of new gear. IOW, the controller code could be merged into the in-core monitor rather than have to be buttembled in with a forced reboot of the system.
Now think about a system failure in an airplane, you automobile, the power company, your house system that governs security, heating, and other maintenance. I disagree so whole-heartedly with you about SMP being a luxury.
When the electric power company wants to bring another dynamo on-line to generate power, is it acceptable to you that they completely shut down the power grid just to add a new bolt?
It is the only way to provide delivery of continuous computing services.
On Thu, 02 Mar 06 13:09:27 GMT No it's not. The most common way of providing delivery of non stop continuous computing services these days is mbuttive redundancy...
when were were doing ha-cmp ... we talked to the 1-800 people. they had five-nines requirement (5 minutes downtime per year) ... and were using a platform that was hardware fault tolerant. however...
We disagree. There is no other way other than a perfect world. Sheesh! Maintaining a computer system should be as easy as maintaining one horse stall.
BAH
IBM 610 workstation computer 3433
Alt Folklore Computers from Newsgroups