From: Penang on
There is an article on OS architecture design truly for the Multi-Core
systems that we are using today.

http://www.networkworld.com/news/2010/031910-multicore-requires-os-rework-windows.html

I'll leave you some quotes from the article for you guys to think
over.

Maybe Linux could use a retooling, from the ground up, as well.

"The current approach to harnessing the power of multicore processors
is complicated and not entirely successful. The key may not be in
throwing more energy into refining techniques such as parallel
programming, but rather rethinking the basic abstractions that make up
the operating systems model.

Today's typical desktop computer runs multiple programs at once,
playing music while the user writes an e-mail and surfs the Web, for
instance. The problem is today's desktop programs don't use the
multiple cores efficiently enough.

Perhaps a better way to deal with multiple cores is to rethink the way
operating systems handle these processors.

The current architecture of operating systems is based on a number of
different abstractions, when we wanted multiple programs to run on a
single processor, the CPU time was sliced up into processes, giving
each application the illusion that it was running on a dedicated CPU.

With all these virtual CPUs, however, come struggles over who gets the
attention of the real CPU. The overhead of switching between all these
CPUs starts to grow to the point where responsiveness suffers,
especially when multiple cores are introduced.

But with Intel and AMD predicting that the core count of their
products will continue to multiply, the OS community may be safe in
jettisoning abstractions such as user mode and kernel mode.

With many-core, CPUs could become CPUs again. If we get enough of
them, maybe we can start to hand them out to individual programs.

In this approach, the operating system would no longer resemble the
kernel mode of today's OSes, but rather act more like a hypervisor. A
concept from virtualization, a hypervisor acts as a layer between the
virtual machine and the actual hardware.

The programs themselves would take on many of the duties of resource
management. The OS could assign an application a CPU and some memory,
and the program itself, using metadata generated by the compiler,
would best know how to use these resources."
From: peterwn on
On Mar 22, 10:11 pm, Penang <kalamb...(a)gmail.com> wrote:
> There is an article on OS architecture design truly for the Multi-Core
> systems that we are using today.
>
> http://www.networkworld.com/news/2010/031910-multicore-requires-os-re...
>
> I'll leave you some quotes from the article for you guys to think
> over.
>
> Maybe Linux could use a retooling, from the ground up, as well.
>

See Peter Kohlmann's comment on another thread:
"You mean multicore like the suercomputers running under linux right
now?
You know, those with *thousands* of cores?"

From: The Natural Philosopher on
peterwn wrote:
> On Mar 22, 10:11 pm, Penang <kalamb...(a)gmail.com> wrote:
>> There is an article on OS architecture design truly for the Multi-Core
>> systems that we are using today.
>>
>> http://www.networkworld.com/news/2010/031910-multicore-requires-os-re...
>>
>> I'll leave you some quotes from the article for you guys to think
>> over.
>>
>> Maybe Linux could use a retooling, from the ground up, as well.
>>
>
> See Peter Kohlmann's comment on another thread:
> "You mean multicore like the suercomputers running under linux right
> now?
> You know, those with *thousands* of cores?"
>
Any piece of constantly developed software benefits from a total rewrite
every ten years or so.

But who has the time to do it?
From: owl on
In comp.os.linux.advocacy The Natural Philosopher <tnp(a)invalid.invalid> wrote:
> peterwn wrote:
>> On Mar 22, 10:11 pm, Penang <kalamb...(a)gmail.com> wrote:
>>> There is an article on OS architecture design truly for the Multi-Core
>>> systems that we are using today.
>>>
>>> http://www.networkworld.com/news/2010/031910-multicore-requires-os-re...
>>>
>>> I'll leave you some quotes from the article for you guys to think
>>> over.
>>>
>>> Maybe Linux could use a retooling, from the ground up, as well.
>>>
>>
>> See Peter Kohlmann's comment on another thread:
>> "You mean multicore like the suercomputers running under linux right
>> now?
>> You know, those with *thousands* of cores?"
>>
> Any piece of constantly developed software benefits from a total rewrite
> every ten years or so.
>
> But who has the time to do it?

Not to worry...

http://www.drury.edu/ess/Culture/indian.htm
<quote>
The Indian conception of time is very different from what the Western
mind regards as intuitively obvious. In Indian thought, time, like other
phenomena, is conceived statically rather than dynamically. It is, of
course, recognized that the things of this world are always moving and
changing. But the substance of things is seen as basically unchanging,
its underlying reality unaffected by the ceaseless flux. The Indian does
not concede that we never step into the same river twice; he directs
our attention not to the flow of water but to the river itself, the
unchanging universal. Indian thought places a high value on universality,
and the connection between this, and the static conception of phenomena,
is of course not accidental. "The one remains, the many change and flee."
</quote>

From: Hadron on
peterwn <pmilne29(a)gmail.com> writes:

> On Mar 22, 10:11 pm, Penang <kalamb...(a)gmail.com> wrote:
>> There is an article on OS architecture design truly for the Multi-Core
>> systems that we are using today.
>>
>> http://www.networkworld.com/news/2010/031910-multicore-requires-os-re...
>>
>> I'll leave you some quotes from the article for you guys to think
>> over.
>>
>> Maybe Linux could use a retooling, from the ground up, as well.
>>
>
> See Peter Kohlmann's comment on another thread:
> "You mean multicore like the suercomputers running under linux right
> now?
> You know, those with *thousands* of cores?"

Those "super computers" use, err, a re-tooled Linux. Surely you're not
so dim as to not realise that. Hint : if Koehlmann mentions something
even remotely technical treat it like a turd on your doorstep. Do not
embrace it and wave it around.