From: Vivek Goyal <vgoyal@redhat.com>
To: Tejun Heo <tj@kernel.org>
Cc: lizefan@huawei.com, axboe@kernel.dk,
containers@lists.linux-foundation.org, cgroups@vger.kernel.org,
linux-kernel@vger.kernel.org, ctalbott@google.com,
rni@google.com
Subject: Re: [PATCH 07/12] cfq-iosched: implement hierarchy-ready cfq_group charge scaling
Date: Mon, 17 Dec 2012 16:27:36 -0500 [thread overview]
Message-ID: <20121217212736.GB13691@redhat.com> (raw)
In-Reply-To: <20121217211738.GD1844@htj.dyndns.org>
On Mon, Dec 17, 2012 at 01:17:38PM -0800, Tejun Heo wrote:
> Hello,
>
> On Mon, Dec 17, 2012 at 03:53:18PM -0500, Vivek Goyal wrote:
> > On Fri, Dec 14, 2012 at 02:41:20PM -0800, Tejun Heo wrote:
> > > Currently, cfqg charges are scaled directly according to cfqg->weight.
> > > Regardless of the number of active cfqgs or the amount of active
> > > weights, a given weight value always scales charge the same way. This
> > > works fine as long as all cfqgs are treated equally regardless of
> > > their positions in the hierarchy, which is what cfq currently
> > > implements. It can't work in hierarchical settings because the
> > > interpretation of a given weight value depends on where the weight is
> > > located in the hierarchy.
> >
> > I did not understand this. Why the current scheme will not work with
> > hierarchy?
>
> Because the meaning of a weight changes depending on where the weight
> exists in the hierarchy?
>
> > While we calculate the vdisktime, this is calculated with the help
> > of CFQ_DEFAULT_WEIGHT and cfqg->weight. So we scale used time slice
> > in proportion to CFQ_DEFAULT_WEIGTH/cfqg->weight. So higher the weight
> > lesser the charge and cfqg gets scheduled again faster and lower the
> > weight, higher the vdisktime and cfqg gets scheduled less frequently.
> >
> > As every cfqg does the same thing on service tree, they automatically
> > get fair share w.r.t their weight.
> >
> > And this mechanism should not be impacted by the hierarchy because we
> > have a separate service tree at separate level. This will not work
> > only if you come up with one compressed tree and then weights will
> > have to be adjusted. If we have a separate service tree in each group
> > then it should work just fine.
>
> Why would you create N service trees when you can almost trivially use
> one by calcualting the effective weight? You would have to be
> adjusting all trees above whenever something changes in a child.
One of the reasons I can think is accuracy. If a task/group is added to
a service tree, it mostly does not change the fraction of parent and
does not change the fraction of parent's sibling.
By making everything flat any addition/removal of an entity changes
fraction of everything on the tree.
Not that I am bothered about it because we do not focus that strictly
on fairness. So I would not care about it.
What I do care about is atleast being able to read and understand the
code easily. Right now, it is hard to understand. I am still struggling
to wrap my head around it.
For example, while adding a group to service tree we calculate
cfqg->vfaction as follows.
vfr = vfr * pos->leaf_weight / pos->level_weight;
and then
vfr = vfr * pos->weight / parent->level_weight;
cfqg->vfraction = max_t(unsigned, vfr, 1)
If cfqg->vfraction is about cfqg then why should we take into account
leaf_weight and level_weight. We should be just worried about pos->weight
and parent->level_weight and that should determine vfaction of cfqg.
Thanks
Vivek
next prev parent reply other threads:[~2012-12-17 21:27 UTC|newest]
Thread overview: 40+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-12-14 22:41 [PATCHSET] block: implement blkcg hierarchy support in cfq Tejun Heo
2012-12-14 22:41 ` [PATCH 01/12] blkcg: fix minor bug in blkg_alloc() Tejun Heo
2012-12-17 19:10 ` Vivek Goyal
2012-12-14 22:41 ` [PATCH 02/12] blkcg: reorganize blkg_lookup_create() and friends Tejun Heo
2012-12-17 19:28 ` Vivek Goyal
2012-12-14 22:41 ` [PATCH 03/12] blkcg: cosmetic updates to blkg_create() Tejun Heo
2012-12-17 19:37 ` Vivek Goyal
2012-12-14 22:41 ` [PATCH 04/12] blkcg: make blkcg_gq's hierarchical Tejun Heo
2012-12-17 20:04 ` Vivek Goyal
2012-12-14 22:41 ` [PATCH 05/12] cfq-iosched: add leaf_weight Tejun Heo
2012-12-14 22:41 ` [PATCH 06/12] cfq-iosched: implement cfq_group->nr_active and ->level_weight Tejun Heo
2012-12-17 20:46 ` Vivek Goyal
2012-12-17 21:15 ` Tejun Heo
2012-12-17 21:18 ` Vivek Goyal
2012-12-17 21:20 ` Tejun Heo
2012-12-14 22:41 ` [PATCH 07/12] cfq-iosched: implement hierarchy-ready cfq_group charge scaling Tejun Heo
2012-12-17 20:53 ` Vivek Goyal
2012-12-17 21:17 ` Tejun Heo
2012-12-17 21:27 ` Vivek Goyal [this message]
2012-12-17 21:33 ` Tejun Heo
2012-12-17 21:49 ` Vivek Goyal
2012-12-17 22:12 ` Tejun Heo
2012-12-14 22:41 ` [PATCH 08/12] cfq-iosched: convert cfq_group_slice() to use cfqg->vfraction Tejun Heo
2012-12-14 22:41 ` [PATCH 09/12] cfq-iosched: enable full blkcg hierarchy support Tejun Heo
2012-12-18 18:40 ` Vivek Goyal
2012-12-18 19:10 ` Tejun Heo
2012-12-18 19:16 ` Vivek Goyal
2012-12-18 19:17 ` Tejun Heo
2012-12-14 22:41 ` [PATCH 10/12] blkcg: add blkg_policy_data->plid Tejun Heo
2012-12-14 22:41 ` [PATCH 11/12] blkcg: implement blkg_prfill_[rw]stat_recursive() Tejun Heo
2012-12-14 22:41 ` [PATCH 12/12] cfq-iosched: add hierarchical cfq_group statistics Tejun Heo
2012-12-18 19:11 ` Vivek Goyal
2012-12-18 19:14 ` Tejun Heo
2012-12-18 19:18 ` Vivek Goyal
2012-12-18 19:21 ` Tejun Heo
2012-12-18 19:26 ` Vivek Goyal
2012-12-17 16:52 ` [PATCHSET] block: implement blkcg hierarchy support in cfq Vivek Goyal
2012-12-17 17:38 ` Tejun Heo
2012-12-17 18:50 ` Vivek Goyal
2012-12-17 18:59 ` Tejun Heo
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20121217212736.GB13691@redhat.com \
--to=vgoyal@redhat.com \
--cc=axboe@kernel.dk \
--cc=cgroups@vger.kernel.org \
--cc=containers@lists.linux-foundation.org \
--cc=ctalbott@google.com \
--cc=linux-kernel@vger.kernel.org \
--cc=lizefan@huawei.com \
--cc=rni@google.com \
--cc=tj@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).