From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <vdavydov.dev@gmail.com>
Date: Fri, 24 Aug 2018 11:31:26 +0300
From: Vladimir Davydov <vdavydov.dev@gmail.com>
Subject: Re: [PATCH 11/18] vinyl: use snap_io_rate_limit for initial dump
 bandwidth estimate
Message-ID: <20180824083126.6wjnwkokmyty7fj2@esperanza>
References: <cover.1534432819.git.vdavydov.dev@gmail.com>
 <111debb195d9c9780edef15d311ea94282a4fd60.1534432819.git.vdavydov.dev@gmail.com>
 <20180820112431.GN8716@chai>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20180820112431.GN8716@chai>
To: Konstantin Osipov <kostja@tarantool.org>
Cc: tarantool-patches@freelists.org
List-ID: <tarantool-patches.dev.tarantool.org>

On Mon, Aug 20, 2018 at 02:24:31PM +0300, Konstantin Osipov wrote:
> * Vladimir Davydov <vdavydov.dev@gmail.com> [18/08/16 23:03]:
> > The user can limit dump bandwidth with box.cfg.snap_io_rate_limit to a
> > value, which is less than the current estimate. To avoid stalls caused
> > by overestimating dump bandwidth, we must take into account the limit
> > for the initial guess and forget all observations whenever it changes.
> 
> >  void
> > +vy_quota_reset_dump_bw(struct vy_quota *q, size_t max)
> > +{
> > +	histogram_reset(q->dump_bw_hist);
> > +	q->dump_bw = MIN(VY_DEFAULT_DUMP_BANDWIDTH, max);
> > +}
> > +
> 
> This sounds like vy_quota_set_dump_bw to me, not reset().

I think it's rather 'reset', because it forgets all measurements that
have been observed so far.