Re: [RFC PATCH v2 00/10] mm/damon: support multiple goal-based quota tuning algorithms

[SeongJae Park <sj@kernel.org>]

On Wed, 4 Mar 2026 13:15:29 +0300 Gutierrez Asier <gutierrez.asier@huawei-partners.com> wrote:

> Hi SeongJae!
> 
> Nice idea for dynamic environments.

Thank you :)

> 
> On 3/4/2026 7:41 AM, SeongJae Park wrote:
> > Aim-oriented DAMOS quota auto-tuning uses a single tuning algorithm.
> > The algorithm is designed to find a quota value that should be
> > consistently kept for achieving the aimed goal for long term.  It is
> > useful and reliable at automatically operating systems that have dynamic
> > environments in the long term.
> > 
> > As always, however, no single algorithm fits all.  When the environment
> > has static characteristics or there are control towers in not only the
> > kernel space but also the user space, the algorithm shows some
> > limitations.  In such environments, users want kernel work in a more
> > short term deterministic way.  Actually there were at least two reports
> > [1,2] of such cases.
> > 
> > Extend DAMOS quotas goal to support multiple quota tuning algorithms
> > that users can select.  Keep the current algorithm as the default one,
> > to not break the old users.  Also give it a name, "consist", as it is
> > designed to "consistently" apply the DAMOS action.  And introduce a new
> > tuning algorithm, namely "temporal".  It is designed to apply the DAMOS
> > action only temporally, in a deterministic way.  In more detail, as long
> > as the goal is under-achieved, it uses the maximum quota available.
> > Once the goal is over-achieved, it sets the quota zero.
> 
> I'm not sure "temporal" is the best name for this type of behaviour.

I agree there could be a better name.

> 
> How about "by_score?". For example, "damos_goal_tune_esz_bp_by_score" and
> DAMOS_QUOTA_GOAL_TUNER_BY_SCORE.

And thank you for the suggestion!

But... I don't think "by_score" is much better, because all tuners are assumed
to, and actually do the tuning of the quota based on the score.  Or, maybe you
mean it makes non-zero quota only until the score becomes the goal?  That makes
sense, but again, in a sense, that's same for "consistent" tuner.

Naming is difficult...

I was also thinking about a few more names, but my conclusion after the self
discussion was that some of ambitious names are inevitable here.  Otherwise,
the name will be too long.  I therefore picked the shortest and simplest ones
on my list, which at least contrasts the current two tuners.  I agree that
could still be difficult to understand.  But as long as there is a good
documentation, I think difficult-to-understnd names that encourage users to
read the document is ok and might even be better in some cases.

I'm of course open to other suggestions.


Thanks,
SJ

[...]