Another XFS Fan

[Steven Alligood]

Joseph Hall wrote:
> On Wed, Mar 12, 2008 at 9:36 PM, Gabriel Gunderson <gabe at gundy.org> wrote:
>   
>> On Wed, 2008-03-12 at 21:15 -0600, Stuart Jansen wrote:
>>  > I didn't compare performance because all I cared about was space
>>  > efficiency.
>>
>>  I'm not commenting on the test here (I've never done any testing of my
>>  own and haven't really read up on it), but I thought this sentence was
>>  interesting.  When it comes to disks, there are two things you can't
>>  ignore - disk access is the biggest bottleneck in almost any system and
>>  disks space is getting cheap.
>>
>>  I was caught off guard because I expected to read, "I didn't compare
>>  space efficiency because all I cared about was performance."
>>     
>
> Certainly a valid concern, but I'm kind of poor. Until drive space
> goes from "cheap" to "free", I'm still interested in drive efficiency.
>
> I'm also interested in performance, however. Assuming one had spare
> hardware at hand, such as a classroom full of unused computers, what
> would be a valid series of tests to run to evaluate efficiency?
>
>   
If it is just space you are after, both xfs and reiser will save you 
space at the cost of performance.

The truly great thing about xfs is that you can tune it to whatever you 
want.  Default is space savings, but you can tune it for performance 
savings, and even specifically for the kinds of stuff an individual 
server does.

The price is complexity.  ext3 is generally the best bet for simplicity 
and ease of use, fast recovery, etc.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3241 bytes
Desc: S/MIME Cryptographic Signature
Url : http://plug.org/pipermail/plug/attachments/20080313/547c1cff/smime.bin