Operating System - HP-UX
1826325 Members
3515 Online
109692 Solutions
New Discussion

Filesystem Utilization Thresholds

 
COMAIR IT Department
Occasional Contributor

Filesystem Utilization Thresholds

I remember at one point being told that VxFS performance could be impacted if total filesystem utilization for a mount point creeps beyond 90%. Perhaps this is old information or an urban myth. In any case, I prefer to keep ours below 90% in case something begins to run away to provide growing room.

Can anyone validate what I'm thinking?
2 REPLIES 2
Sanjay_6
Honored Contributor

Re: Filesystem Utilization Thresholds

Hi,

Maybe this link will help,

http://www1.itrc.hp.com/service/cki/docDisplay.do?docLocale=en_US&docId=200000075725020

The itrc doc id is KBRC00015029.

Hope this helps.

Regds
A. Clay Stephenson
Acclaimed Contributor

Re: Filesystem Utilization Thresholds

Refer to this document:

http://docs.hp.com/cgi-bin/fsearch/framedisplay?top=/hpux/onlinedocs/5971-4772/5971-4772_top.html&con=/hpux/onlinedocs/5971-4772/00/00/29-con.html&toc=/hpux/onlinedocs/5971-4772/00/00/29-toc.html&searchterms=performance%7ctuning%7cVxfs&queryid=20041123-134545

but here's the relavant section:
Quote:

Monitoring Free Space



In general, VxFS works best if the percentage of free space in the file system does not get below 10 percent. This is because file systems with 10 percent or more free space have less fragmentation and better extent allocation. Regular use of the df command (see the df_vxfs(1M) manual page) to monitor free space is desirable. Full file systems may have an adverse effect on file system performance. Full file systems should therefore have some files removed, or should be expanded (see the fsadm_vxfs(1M) manual page for a description of online file system expansion).
Unquote

Note the "In general" and translate that to it depends. Suppose for example, this is a filesystem housing Oracle data files that are preallocated and thus the filesystem is static. In that case, there is no reason not to use almost all the capacity.

Even the fragmenting is usually overstated because, IN GENERAL, the improvements to be gained by defragmenting while real are IN GENERAL very small and often very difficult to measure. Again, in general, I'm usually much more concerned about capacity than about performance degradation because one will stop an application in its tracks while the other usually has no measurable impact.


If it ain't broke, I can fix that.