From mboxrd@z Thu Jan 1 00:00:00 1970 From: ludo@gnu.org (Ludovic =?utf-8?Q?Court=C3=A8s?=) Subject: Re: Speed difference: wget vs guix build Date: Wed, 06 May 2015 10:53:50 +0200 Message-ID: <877fsmulwh.fsf@gnu.org> References: <87383aw77m.fsf@mango.localdomain> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Return-path: Received: from eggs.gnu.org ([2001:4830:134:3::10]:39277) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1Ypv5b-0008Ky-5W for guix-devel@gnu.org; Wed, 06 May 2015 04:54:01 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1Ypv5W-0002v3-HT for guix-devel@gnu.org; Wed, 06 May 2015 04:53:59 -0400 Received: from fencepost.gnu.org ([2001:4830:134:3::e]:59999) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1Ypv5W-0002up-EP for guix-devel@gnu.org; Wed, 06 May 2015 04:53:54 -0400 In-Reply-To: <87383aw77m.fsf@mango.localdomain> (Ricardo Wurmus's message of "Wed, 6 May 2015 08:28:13 +0200") List-Id: "Development of GNU Guix and the GNU System distribution." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: guix-devel-bounces+gcggd-guix-devel=m.gmane.org@gnu.org Sender: guix-devel-bounces+gcggd-guix-devel=m.gmane.org@gnu.org To: Ricardo Wurmus Cc: Guix-devel Ricardo Wurmus skribis: > [rwurmus@guix-builder:~] (557) $ time wget http://hydra.gnu.org/nar/h= 1qjkg8bvcdb485q81lw5844szgywpg7-texlive-texmf-2014 > --2015-05-06 08:20:38-- http://hydra.gnu.org/nar/h1qjkg8bvcdb485q81l= w5844szgywpg7-texlive-texmf-2014 > Resolving hydra.gnu.org (hydra.gnu.org)... 18.4.89.46 > Connecting to hydra.gnu.org (hydra.gnu.org)|18.4.89.46|:80... connect= ed. > HTTP request sent, awaiting response... 200 OK > Length: unspecified [application/x-nix-archive] > Saving to: =E2=80=98h1qjkg8bvcdb485q81lw5844szgywpg7-texlive-texmf-20= 14.1=E2=80=99 > > [ <=3D> = ] 761,111,979 4.17MB/s > > The speed ranges from ~4MB/s to 7MB/s with wget. > With guix build, however, I get much less on the same host: > > [rwurmus@guix-builder:/localhome/rwurmus-tmp/guix] (557) $ ./pre-inst= -env guix build texlive-texmf-2014 > The following file will be downloaded: > /gnu/store/h1qjkg8bvcdb485q81lw5844szgywpg7-texlive-texmf-2014 > @ substituter-started /gnu/store/h1qjkg8bvcdb485q81lw5844szgywpg7-tex= live-texmf-2014 /usr/libexec/guix/substitute > found valid signature for '/gnu/store/h1qjkg8bvcdb485q81lw5844szgywpg= 7-texlive-texmf-2014', from 'http://hydra.gnu.org/nar/h1qjkg8bvcdb485q81lw5= 844szgywpg7-texlive-texmf-2014' > downloading `/gnu/store/h1qjkg8bvcdb485q81lw5844szgywpg7-texlive-texm= f-2014' (3391.2 MiB installed)... > http://hydra.gnu.org/nar/h1qjkg8bvcdb485q81lw5844szgywpg7-texlive-tex= mf-2014 124344.0 KiB transferred (321. KiB/s) Beware: wget shows the =E2=80=9Cinstantaneous=E2=80=9D throughput, whereas = Guix shows the average throughput since the beginning. So to be sure, you need to make a full download. Also, please compare not only against =E2=80=98guix substitute=E2=80=99 but also against =E2=80=98guix download=E2=80=99. Regardless, I don=E2=80=99t see anything like that here. Commit c822fb8 may help a bit; could you retry? Commit 7623848 helps for source derivations and =E2=80=98guix download=E2= =80=99 for Guile <=3D 2.0.11, but it shouldn=E2=80=99t make any difference for =E2=80= =98guix substitute=E2=80=99. Thanks for the report, Ludo=E2=80=99.