henry@utzoo.uucp (Henry Spencer) (08/12/88)
In article <19709@tut.cis.ohio-state.edu> mudd-j@pike.cis.ohio-state.edu (John R. Mudd) writes: >>Have you ever used VM/CMS? ... > >... I *liked* VM/CMS from a user point-of-view. The shop I worked at had a >IBM 3090 running VM/XA with 200+ users, and that system was FAST. Not like >some of the delays I've had on some of these Unix-boxes. Of course, they're >not as big as a 3090... Damn right they aren't. Unix on a big Amdahl really blasts along too. Don't confuse hardware with software; of course a really big mainframe is faster than an overgrown mini. -- Intel CPUs are not defective, | Henry Spencer at U of Toronto Zoology they just act that way. | uunet!attcan!utzoo!henry henry@zoo.toronto.edu
allbery@ncoast.UUCP (Brandon S. Allbery) (08/16/88)
As quoted from <1988Aug12.061040.18720@utzoo.uucp> by henry@utzoo.uucp (Henry Spencer): +--------------- | In article <19709@tut.cis.ohio-state.edu> mudd-j@pike.cis.ohio-state.edu (John R. Mudd) writes: | >>Have you ever used VM/CMS? ... | > | >... I *liked* VM/CMS from a user point-of-view. The shop I worked at had a | >IBM 3090 running VM/XA with 200+ users, and that system was FAST. Not like | >some of the delays I've had on some of these Unix-boxes. Of course, they're | >not as big as a 3090... | | Damn right they aren't. Unix on a big Amdahl really blasts along too. | Don't confuse hardware with software; of course a really big mainframe | is faster than an overgrown mini. +--------------- And, of course, if you stick those 200+ users and VM/CMS on a 4300-series processor, you start to long for the overgrown minis. I know from experience; Cleveland State's mainframe was usually far slower than ncoast, which is a decidedly *slow* system (386 systems are faster with the same load). ++Brandon -- Brandon S. Allbery, uunet!marque!ncoast!allbery DELPHI: ALLBERY For comp.sources.misc send mail to ncoast!sources-misc
madd@bu-cs.BU.EDU (Jim Frost) (08/24/88)
In article <1988Aug12.061040.18720@utzoo.uucp> henry@utzoo.uucp (Henry Spencer) writes: |In article <19709@tut.cis.ohio-state.edu> mudd-j@pike.cis.ohio-state.edu (John R. Mudd) writes: |>... I *liked* VM/CMS from a user point-of-view. The shop I worked at had a |>IBM 3090 running VM/XA with 200+ users, and that system was FAST. Not like |>some of the delays I've had on some of these Unix-boxes. Of course, they're |>not as big as a 3090... | |Damn right they aren't. Unix on a big Amdahl really blasts along too. |Don't confuse hardware with software; of course a really big mainframe |is faster than an overgrown mini. Another thing that people miss is that the 3090 architecture is designed to offload almost all of the I/O work to other processors. You send off a command and then forget about it until it's through. Some of the commands are very complex. I've seen some UNIX boxes that do this, too, and they get MUCH better throughput than otherwise. The reasons for this are obvious to even the beginner. Of course the kind of architecture the 3090 has is going to cost more since you have lots of periferal processors, but you get high throughput and that's what people pay for. It's too bad that no one seems to be using this idea on UNIX boxes using cheap periferal processors. Considering some of the hardware I've seen lately, though, it is probably coming. Another thing to consider is the available tools. It's very easy to get a lot of work done on a UNIX machine -- even if the UNIX machine can't blast along -- because it has a LOT of stuff running on it. If a single user can get his work done in 10 minutes using tools available under UNIX and can get it done in 40 minutes under something else without similar tools, you get a much larger user turnover in the same amount of time. This has proved to be the case in my experience with a proprietary OS on the 3090 here (VM/VPS) versus smaller UNIX boxes. Students get the same work done faster. Even though 300 students can't use the hardware at the same time, they don't have to wait so long to use it; it evens out. jim frost madd@bu-it.bu.edu