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

Re: Why Linux, Why Debian

On Sat, 14 Feb 2004 18:40, Andreas Metzler <ametzler@downhill.at.eu.org> 
> Russell Coker <russell@coker.com.au> wrote:
> > On Sat, 14 Feb 2004 07:45, Philip Brown <phil@bolthole.com> wrote:
> >> If someone is specifically interested in
> >> "how does solaris compare with linux", I would suggest finding a
> >> SCSI-based x86 machine, and doing the comparison on that box.
> >
> > Why SCSI?  For the same money IDE will significantly outperform SCSI.
> [...]
> Because SCSI-disk turn faster their access-time is smaller. Their
> data-transfer-rate is not bigger, though.
> There is no technical reason for this, it is simply caused by the fact
> that manufacturers position scsi as high-end server product. There
> are disks with "SCSI-type" hardware (10000RPM) and IDE interface (WD
> raptor) but they are not cheaper than equivalent SCSI ones.


According to my local computer shop 10,000rpm IDE disks are cheaper than 
10,000rpm SCSI disks, and bigger too.

Faster rotational speed gives higher bulk transfer rate if there are the same 
number of sectors per track.  Run zcav (part of Bonnie++) on some disks and 
you'll notice that higher rotational speed almost always gives a 
proportionally higher bulk read speed if the capacity is the same.

Many disks with a higher rotational speed have a smaller capacity so they may 
have less sectors per track and may not deliver better bulk transfer rates.  
As a rule of thumb, if you want good performance then you want big capacity.  
Using the first 20G of a 80G disk will give much better performance than 
using all of a 20G disk.

Let's move the discussion to debian-isp if we are going to get into the 
details of disk performance.

http://www.coker.com.au/selinux/   My NSA Security Enhanced Linux packages
http://www.coker.com.au/bonnie++/  Bonnie++ hard drive benchmark
http://www.coker.com.au/postal/    Postal SMTP/POP benchmark
http://www.coker.com.au/~russell/  My home page

Reply to: