On 8/10/05, Leandro Guimaraes Faria C. Dutra <ldutra_at_toyota.com.br> wrote:
>
> > The impact that the actual o/s has on performance is negligble
> > when looking at issues such as configuration, software design,
> > implementation and so on.
>
> OK, but that does not preclude the possibility of a specific
> application depending on something that was badly implemented in a
> particular OS. And if that particular OS already makes part in a
> low-quality culture, the chances of hitting particularly badly implemented
> components in the platform is quite high.
For example directIO on Linux filesystems perhaps? Certainly I view the
Linux culture as shown on /. to be low-quality. Something tells me that
wasn't quite what you had in mind though.
Personally I'm with the earlier contributor who suggested that the raw
performance of an OS is a poor factor to make the driving force behind OS
decisions. Factors such as the available skillsets of your staff (and
candidates in your geographical area), support from both the OS and
application vendors, hardware support and so on will be more important it
seems to me. I'd argue that many of the same factors apply to rdbms choice
these days as well.
--
> Niall Litchfield
> Oracle DBA
> http://www.niall.litchfield.dial.pipex.com
--
http://www.freelists.org/webpage/oracle-l
Received on Wed Aug 10 2005 - 09:33:28 CDT
