From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Date: Mon, 4 Jun 2018 12:55:43 +0300 From: Vladimir Davydov Subject: Re: [PATCH 2/2] vinyl: apply box.cfg.snap_io_rate_limit to dump/compaction Message-ID: <20180604095543.mmbspvicrgeyleyh@esperanza> References: <0686cb69b36e79d9646fb44f2923a73a527ab0f9.1527605229.git.vdavydov.dev@gmail.com> <20180601175628.GB15567@atlas> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20180601175628.GB15567@atlas> To: Konstantin Osipov Cc: tarantool-patches@freelists.org List-ID: On Fri, Jun 01, 2018 at 08:56:28PM +0300, Konstantin Osipov wrote: > * Vladimir Davydov [18/05/29 18:20]: > > > diff --git a/test/vinyl/snap_io_rate.test.lua b/test/vinyl/snap_io_rate.test.lua > > new file mode 100644 > > index 00000000..836bf537 > > --- /dev/null > > +++ b/test/vinyl/snap_io_rate.test.lua > > @@ -0,0 +1,38 @@ > > +fiber = require('fiber') > > +digest = require('digest') > > +test_run = require('test_run').new() > > + > > +MB = 1024 * 1024 > > +TUPLE_SIZE = 1024 > > +TUPLE_COUNT = 100 > > This test runs for 3.8 seconds on my laptop. > > You only need to run it for 0.2 seconds, definitely 1 second to > test the rate. xlog_write() throttles after writing a file for one second. So in order to make sure throttling works as expected, we need to run a test for a few seconds. To speed up the test, we need to rewrite the way throttling works, but I don't think it's worth it. > > Sorry for being painfully finicky about this, but I run tests > every day. > > If you hate this request, don't bother, this is a standalone test > and can run in parallel. Exactly.