Tarantool development patches archive
 help / color / mirror / Atom feed
From: Vladimir Davydov <vdavydov.dev@gmail.com>
To: Konstantin Osipov <kostja@tarantool.org>
Cc: tarantool-patches@freelists.org
Subject: Re: [PATCH 8/8] vinyl: keep track of thread pool idle ratio
Date: Thu, 6 Sep 2018 14:59:57 +0300	[thread overview]
Message-ID: <20180906115957.pjb5t4ug2wcsv7rc@esperanza> (raw)
In-Reply-To: <20180906105742.GK8205@chai>

On Thu, Sep 06, 2018 at 01:57:42PM +0300, Konstantin Osipov wrote:
> * Vladimir Davydov <vdavydov.dev@gmail.com> [18/09/06 13:53]:
> > > > > > To understand whether the disk is fully utilized or can still handle
> > > > > > more compaction load and make right decisions regarding transaction
> > > > > > throttling, we need a metric that would report how much time worker
> > > > > > threads spent being idle. So this patch adds two new metrics to global
> > > > > > statistics, disk.dump_idle_ratio and compact_idle_ratio, which show how
> > > > > > much time dump threads and compaction threads were idle, respectively.
> > > > > > The metrics are updated using the following formula:
> > > > > > 
> > > > > >                        idle_time
> > > > > >   idle_ratio = --------------------------
> > > > > >                dump_period * worker_count
> > > > > 
> > > > > I don't understand the formula. There can be many workers.
> > > > > Is idle time measured per worker or per entire pool? 
> > > > > 
> > > > > If it is measured per entire pool, how is idle time calculated if
> > > > > some workers are busy and some not?
> > > > 
> > > > It is measured for entire pool - note that I divide the result by
> > > > worker_count. E.g. if there were two workers and one of them were
> > > > busy all the time between two last dumps while another were idle,
> > > > idle_ratio would be 0.5.
> > > 
> > > This looks imprecise. Why not measure idle time of each worker and
> > > then even it out over the total number of workers?
> > 
> > That's exactly what I do. I maintain the idle time for each worker
> > thread and then divide it by the total number of workers.
> 
> This is not what you've written in the comment, though.

It is. Quoting the commit message:

} The metrics are updated using the following formula:
} 
}                        idle_time
}   idle_ratio = --------------------------
}                dump_period * worker_count
} 
} where idle_time is the total amount of time workers were idle between
} the last two dumps, dump_period is the time that passed between the last
} two dumps, worker_count is the number of workers in the pool.


> > > Besides, once again, how do you define the window over which you
> > > measure?
> > I use dump period for the window, i.e. time between two subsequent
> > memory dumps. Respectively, I update the idle_ratio after each memory
> > dump. I'm planning to update throttle rate limit there too.
> 
> This choice of time window looks arbitrary. Why do you think it's
> a good choice? 

Dump period seems to be the only reasonable choice for a minimal time
window characterising a vinyl workload. If we choose a smaller window,
then dump_idle_ratio will jump from 0 when dump is inactive to 1 when
dump is in progress, which isn't very convenient.

At the same time, if we want to accumulate idle time statistics over a
longer period, we can still do that by averaging idle_ratio.

  reply	other threads:[~2018-09-06 11:59 UTC|newest]

Thread overview: 39+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-09-02 20:18 [PATCH 0/7] vinyl: improve stats for throttling Vladimir Davydov
2018-09-02 20:18 ` [PATCH 1/7] vinyl: fix accounting of secondary index cache statements Vladimir Davydov
2018-09-02 22:26   ` [tarantool-patches] " Konstantin Osipov
2018-09-02 20:18 ` [PATCH 2/7] vinyl: add global memory stats Vladimir Davydov
2018-09-02 22:27   ` [tarantool-patches] " Konstantin Osipov
2018-09-02 22:27   ` Konstantin Osipov
2018-09-03  8:10     ` Vladimir Davydov
2018-09-02 20:18 ` [PATCH 3/7] vinyl: add global disk stats Vladimir Davydov
2018-09-02 22:30   ` [tarantool-patches] " Konstantin Osipov
2018-09-02 20:18 ` [PATCH 4/7] vinyl: fix force compaction logic Vladimir Davydov
2018-09-02 20:18 ` [PATCH 5/7] vinyl: update compact priority usual way on range split/coalesce Vladimir Davydov
2018-09-02 20:18 ` [PATCH 6/7] vinyl: keep track of compaction queue length and debt Vladimir Davydov
2018-09-02 20:19 ` [PATCH 7/7] vinyl: keep track of disk idle time Vladimir Davydov
2018-09-04 11:54   ` Vladimir Davydov
2018-09-04 17:23     ` Vladimir Davydov
2018-09-04 17:23       ` [PATCH 1/8] vinyl: add helper to check whether dump is in progress Vladimir Davydov
2018-09-06  7:33         ` Konstantin Osipov
2018-09-04 17:23       ` [PATCH 2/8] vinyl: don't use mempool for allocating background tasks Vladimir Davydov
2018-09-06  7:33         ` Konstantin Osipov
2018-09-04 17:23       ` [PATCH 3/8] vinyl: factor out worker pool from scheduler struct Vladimir Davydov
2018-09-06  7:34         ` Konstantin Osipov
2018-09-04 17:23       ` [PATCH 4/8] vinyl: move worker allocation closer to task creation Vladimir Davydov
2018-09-06  7:35         ` Konstantin Osipov
2018-09-04 17:23       ` [PATCH 5/8] vinyl: use separate thread pools for dump and compaction tasks Vladimir Davydov
2018-09-06  7:37         ` Konstantin Osipov
2018-09-06  9:48           ` Vladimir Davydov
2018-09-06 10:32             ` Konstantin Osipov
2018-09-04 17:23       ` [PATCH 6/8] vinyl: zap vy_worker_pool::idle_worker_count Vladimir Davydov
2018-09-06  7:38         ` Konstantin Osipov
2018-09-04 17:23       ` [PATCH 7/8] vinyl: don't start scheduler fiber until local recovery is complete Vladimir Davydov
2018-09-06  7:39         ` Konstantin Osipov
2018-09-04 17:23       ` [PATCH 8/8] vinyl: keep track of thread pool idle ratio Vladimir Davydov
2018-09-06  7:49         ` Konstantin Osipov
2018-09-06  8:18           ` Vladimir Davydov
2018-09-06 10:26             ` Konstantin Osipov
2018-09-06 10:52               ` Vladimir Davydov
2018-09-06 10:57                 ` Konstantin Osipov
2018-09-06 11:59                   ` Vladimir Davydov [this message]
2018-09-09 11:41 ` [PATCH 0/7] vinyl: improve stats for throttling Vladimir Davydov

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=20180906115957.pjb5t4ug2wcsv7rc@esperanza \
    --to=vdavydov.dev@gmail.com \
    --cc=kostja@tarantool.org \
    --cc=tarantool-patches@freelists.org \
    --subject='Re: [PATCH 8/8] vinyl: keep track of thread pool idle ratio' \
    /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

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox