On Fri, 2 Jun 2000, Mike Porter wrote:
> On Fri, 2 Jun 2000, Markus Pfeiffer wrote:
>
> > Sasi Peter wrote:
> > >
> > > On Thu, 1 Jun 2000, Gregory Maxwell wrote:
> > >
> > > > About the only think I can think of is turning off CPUs on a SMP
> > > > laptop for powermanagement.. :P So unless someone from transmeta speaks
> > > > up about a quad cpu laptop this just sounds like a feature looking for a
> > > > need. (i.e. worthless cruft)
> > >
> > > No!
> > >
> > > Think of a fan failing on one of the CPUs, in the middle of a veri
> > > important computing task. If you can umount that CPU, it will not burn and
> > > crash from overheat, and your process can also finish off.
> > >
> > > Thinking of HA, when desinging the future of linux is a must...
> >
> > ehemmm,
> > I think the CPU will burn even if it is umounted ... and a fan failing
> > ...
> > Good, if you take a very cheap one, but if one does high performance
> > computing
> > one doesn´t use crappy equipment, and even if a failure of expensive
> > equipment
> > is possible, I think the scheduler changes are too expensive ( of the
> > speeds point of
> > view ... ). People at Be seem to be playing people...
>
> Mainframe OSes (IBM MVS, now OS/390) can 'vary' off CPUs. It's not
> something I've done very often, but it can be handy. We had a CPU
> fail one time. The address space (process) running on that
> processor failed, but the operating system kept running, along with
> the 500 other address spaces. Being able to configure CPUs on and
> offline is something you want if you are running a high
> availability system. Detecting failed CPUs and automatically
> configuring them offline is also useful, but does a typical
> consumer PC with Pentium IIIs have any way to communicate the
> notion of a failed processor to the operating system? Probably
> not. Is IBM releasing enough details about their hardware to be
> able to implement failed CPU detection? I'll bet not...
However, it is the sort of feature that will be needed in upcoming
machine. Imagine a 64-way NUMA machine with calcuations that might run for
weeks. The ability to hot-plug cpu field replaceable units would be a
great selling point. As Linux moves into different hardware spaces, the
operating system will have to adapt.
Another example is support of ECC memory.
John Alvord
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Wed Jun 07 2000 - 21:00:15 EST