From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:470:142:3::10]:39434) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1jSjXe-0001Do-W5 for guix-patches@gnu.org; Sun, 26 Apr 2020 11:50:03 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.90_1) (envelope-from ) id 1jSjXe-0004tH-HV for guix-patches@gnu.org; Sun, 26 Apr 2020 11:50:02 -0400 Received: from debbugs.gnu.org ([209.51.188.43]:50647) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_128_GCM_SHA256:128) (Exim 4.90_1) (envelope-from ) id 1jSjXe-0004se-5A for guix-patches@gnu.org; Sun, 26 Apr 2020 11:50:02 -0400 Received: from Debian-debbugs by debbugs.gnu.org with local (Exim 4.84_2) (envelope-from ) id 1jSjXe-00031U-3H for guix-patches@gnu.org; Sun, 26 Apr 2020 11:50:02 -0400 Subject: [bug#39258] benchmark search: default vs v2 vs v3 Resent-Message-ID: From: Ludovic =?UTF-8?Q?Court=C3=A8s?= References: Date: Sun, 26 Apr 2020 17:49:06 +0200 In-Reply-To: (zimoun's message of "Sun, 26 Apr 2020 05:54:21 +0200") Message-ID: <87lfmi2qod.fsf@gnu.org> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: guix-patches-bounces+kyle=kyleam.com@gnu.org Sender: "Guix-patches" To: zimoun Cc: Arun Isaac , Pierre Neidhardt , 39258@debbugs.gnu.org Hi, zimoun skribis: > 1) around 25 seconds added to "guix pull"... but I am more than often > waiting around 10 minutes when pulling. > 2) the speedup is clear: more than 2x. Nice! It does seem like Arun=E2=80=99s v3 (or maybe even v2) would work nicely. > The question is the tradeoff between: the slowdown of pull vs the > speedup of search. What is acceptable? That=E2=80=99s only one criterion among others. I hear the argument that 2= 5s is =E2=80=9Cnothing=E2=80=9D compared to the rest, but it=E2=80=99s really a t= radeoff. Like, if I spent a day optimizing =E2=80=98guix pull=E2=80=99 and managed to save 25s,= I would find it nice. :-) > $ time guix pull -C ~/.config/guix/default-channels.scm It also depends on what=E2=80=99s in that file, of course. > Then, if I pull back from 8cf6d15 to '--commit=3Da357849f5b' then it take= s: > > real 2m13.693s > user 1m37.418s > sys 0m0.666s For me: --8<---------------cut here---------------start------------->8--- $ guix describe Generacio 139 Apr 13 2020 21:50:08 (nuna) guix bad368b repository URL: https://git.savannah.gnu.org/git/guix.git branch: master commit: bad368b0d794689f3a8a11b58f1ea4987938682e $ time guix pull -p /tmp/test --commit=3Dbad368b0d794689f3a8a11b58f1ea49879= 38682e Updating channel 'guix' from Git repository at 'https://git.savannah.gnu.or= g/git/guix.git'... Building from this channel: guix https://git.savannah.gnu.org/git/guix.git bad368b [...] real 0m57.916s user 1m1.017s sys 0m0.609s --8<---------------cut here---------------end--------------->8--- (On a 2.6=C2=A0GHz i7 though.) > Well, let remove the profiles and garbage collect the index files: > > rm /tmp/default /tmp/v{2,3}* > guix gc -D \ > /gnu/store/g5c08vqsv31nkn2r0hr32dbrkhf3cvd8-guix-package-cache \ > /gnu/store/8xbzhn81hmshagbgazmnr7xfps1cdsa3-guix-package-search-index \ > /gnu/store/8j78b5c4ddic21gcx7wpbq2akjn7x7mr-guix-package-metadata-cache Could you do, for v2 and v3: time guix build /gnu/store/=E2=80=A6-package-metadata-cache.drv --check ? That we=E2=80=99ll give us the exact cost of that part. It=E2=80=99ll be i= nteresting especially in the Xapian case, which we expected to be higher. Thanks for the insightful benchmarks! Ludo=E2=80=99.