[ale] file serving

[David Hamm]

Yep.  But this is a special case.  The device the workstations are driving 
can't move any faster than if the workstations were connected via 10 megabit.  
The files are also cached locally prior to being driven to the output device.   
I can't imagine any reason a 1.2 ghz celeron wouldn't perform fine.  But it 
never hurts to ask the folks on this list.  

On Friday 20 February 2004 03:26 pm, Jeff Hubbs wrote:
> Are you saying all the workstations are also gigabit?
>
> - Jeff
>
> On Fri, 2004-02-20 at 14:48, David Hamm wrote:
> > Do you think using gigabit over ethernet would have any impact on this? 
> > The link between the servers and workstations is gigabit.
> >
> > On Friday 20 February 2004 02:32 pm, Jeff Hubbs wrote:
> > > David -
> > >
> > > My experience suggests not.  100base-T is going to be the limiting
> > > factor; you can watch top or whatever when you're blasting files down
> > > via Samba and see that.  However, if you do a lot of tarball
> > > manipulation or md5sum checking on your fileserver, then it may well
> > > matter.
> > >
> > > - Jeff
> > >
> > > On Fri, 2004-02-20 at 14:24, David Hamm wrote:
> > > > Does anyone know at what point CPU power doesn't help in file serving
> > > > via SMB. Can you get comparable performance from a PIII as you would
> > > > a P4. Is a Celeron slower at serving files than a comparable speed
> > > > P4?  Any links or suggestions are welcome.  Thanks.
> > > >
> > > > _______________________________________________
> > > > Ale mailing list
> > > > Ale at ale.org
> > > > http://www.ale.org/mailman/listinfo/ale
> > >
> > > _______________________________________________
> > > Ale mailing list
> > > Ale at ale.org
> > > http://www.ale.org/mailman/listinfo/ale
> >
> > _______________________________________________
> > Ale mailing list
> > Ale at ale.org
> > http://www.ale.org/mailman/listinfo/ale
>
> _______________________________________________
> Ale mailing list
> Ale at ale.org
> http://www.ale.org/mailman/listinfo/ale