On Fri, Mar 09, 2001 at 04:10:15PM +0700, Robert Elz wrote:
> [...]
> There were some common tests at CSRG they used for this kind of testing,
> though what they were I no longer recall.  Personally I just used "normal
> work" on the system, which at the time was heavy enough, and roughly
> consistent enough, to give reliable results - that is, just running the
> system with quotas enabled for a day, measuring how much time was spent
> handling the quota stuff (including how much I/O was generated for qouta
> activities).  The numbers were always small enough to not be a worry.

As an addendum to this, significant work was done by CSRG to make 4.3
performance better than 4.2 - the namei() transfer stuff, et al.  (At
the time, CSRG were saying that even minor numbers were for
functionality additions, odd minor numbers were for performance
improvements and tuning).  I strongly believe that if the quota
functionality had added any kind of overhead, it would have been noted
at the time.  It was not.

Regards,
Alistair