DragonFly BSD
DragonFly users List (threaded) for 2005-03
[Date Prev][Date Next]  [Thread Prev][Thread Next]  [Date Index][Thread Index]

Re: Fwd: Re: Differences between constructive anddestructivecriticism

From: EM1897@xxxxxxx
Date: Sat, 12 Mar 2005 18:54:12 -0500

In a message dated 3/12/2005 6:20:23 PM Eastern Standard Time, Chris Pressey <cpressey@xxxxxxxxxxxxxxx> writes:

>On Sat, 12 Mar 2005 18:08:05 -0500
>EM1897@xxxxxxx wrote:
>> In a message dated 3/12/2005 4:02:33 PM Eastern Standard Time, Chris
>> Pressey <cpressey@xxxxxxxxxxxxxxx> writes:
>> >On Sat, 12 Mar 2005 14:02:26 -0500
>> >EM1897@xxxxxxx wrote:
>> >
>> >> > Chris Pressey babbled:
>> >> 
>> >> >On Sat, 12 Mar 2005 13:20:20> EM1897@xxxxxxx wrote:
>> >> >
>> >> >> The purpose of an O/S is to utilize hardware. If you think
>> >> >> that hardware is irrelevant, then you really aren't paying
>> >> >> attention. The entire point of FreeBSD 5 and dragonfly is
>> >> >> that uniprocessor has hit a wall, and the freebsd 4.x
>> >> >> design is not multiprocessor friendly, so it needs to be
>> >> >> redesigned to utilize multi-core, MP architectures. Its
>> >> >> always about performance, and the hardware available is
>> >> >> the entire point.
>> >> >
>> >> >How do you reconcile this with your earlier, clearly non-MP view?
>> >> >
>> >> >"If you learn to read english you'll know that uniprocessor is a
>> >> >basic premise."
>> >> 
>> >> There are currently no MP O/S implementations that run faster
>> >> than FreeBSD 4.x on a uniprocessor system, so thats what
>> >> I use. There is no point in investing in expensive hardware
>> >> that is slower than a uniprocessor solution.
>> >
>> >Wait - so if UP is still faster than MP, how can it have possibly
>> >"hit a wall"?
>> Is that a trick question? Surely you're not serious?
>No trick; I honestly have a hard time following you.  Please explain?
>> You know Chris, I listen to politicians complaining about 
>> the disparity between the rich and the poor, and then I 
>> encounter someone like you who apparently has no ability
>> to grasp even the simplist of concepts, and I'm amazed that
>> someone like you can even get a job at all, much less 
>> compete with competent technical people. 
>Please understand that I'm an application programmer at heart.  (Which
>may help explain why I work on the installer and not on the kernel.)

Processor speeds have "hit a wall", which means that they
can't get much faster. So the best way to get more 
performance is to increase support for MP. My point was
that current 'BSD OSes don't utilize MP very well, so 
the best price/performance is with UP. I find the empirical
evidence somewhat undeniable, but apparently the notion
is irritating to people. Why do I care if other people
waste their money? That, I can't answer. 

The "other" point, was that if I were to accept the premise
that opterons are empirically faster than P4s, you can't
possibly get better performance on a PCI bus than you do
on a PCI-X bus, so quoting opteron MBs that don't have a
PCI-X bus is not quoting a comparable system, assuming
that networking is an important component of why you 
are buying a system in the first place. It seems that
there are many that find this notion irritable as well, 
even though all of the empirical evidence is that running
NICs on a faster bus substantially decreases the CPU load
of the system. Now if someone actually had evidence to the
contrary, it might be an interesting discussion. But what
we have is a bunch of evidence in the form of easily 
duplicatable measurements, and a mob with no evidence at
all, and apparently no experience even using the bus 
since they think its bleeding edge (even though its
been the dominant but for many years, telling me I'm crazy. 
So the only possible conclusion is that there isn't any intelligent life amongst the mob members. Since both
Matt and Joerge seem to consider a fast bus  a "strange
requirement", one has to wonder how dragonfly will ever
be optimized for networking?

Hope that clarifies things.

[Date Prev][Date Next]  [Thread Prev][Thread Next]  [Date Index][Thread Index]