Call for Discussion: New Project: Skara -- sample repacked git repo of JDK 12 sources now available

[Martijn Verburg]

Would be interesting to compare GitHub uptime vs OpenJDK hosted uptime in
the past year (honestly couldn’t guess at which has more 9’s).

On Mon, 13 Aug 2018 at 15:38, Mario Torre <neugens.limasoftware at gmail.com>
wrote:

> Nice, however I would not like to rely on GitHub. What if it’s down in the
> middle of a security errata?
>
> Cheers,
> Mario
> On Mon 13. Aug 2018 at 13:40, Remi Forax <forax at univ-mlv.fr> wrote:
>
> > from an island on the atlantic coast of France:
> >
> > time git clone https://github.com/Project-Skara/jdk.git
> > Cloning into 'jdk'...
> > remote: Counting objects: 881253, done.
> > remote: Compressing objects: 100% (160363/160363), done.
> > remote: Total 881253 (delta 665729), reused 881253 (delta 665729),
> > pack-reused 0
> > Receiving objects: 100% (881253/881253), 258.61 MiB | 1.49 MiB/s, done.
> > Resolving deltas: 100% (665729/665729), done.
> > Checking out files: 100% (66258/66258), done.
> >
> > real    3m22.779s
> > user    1m9.630s
> > sys     0m12.783s
> >
> > hg clone for me is 24 minutes (when there is no connection reset by
> peer).
> >
> > The fact that github as mirrors in Europe is a huge win.
> >
> > Rémi
> >
> > ----- Mail original -----
> > > De: "Aleksey Shipilev" <shade at redhat.com>
> > > À: "joe darcy" <joe.darcy at oracle.com>, "discuss" <
> > discuss at openjdk.java.net>
> > > Envoyé: Lundi 13 Août 2018 09:51:22
> > > Objet: Re: Call for Discussion: New Project: Skara -- sample repacked
> > git repo of JDK 12 sources now available
> >
> > > On 08/13/2018 07:26 AM, joe darcy wrote:
> > >> PPS A repacked sample git repo, as alluded to earlier in this thread
> > and as
> > >> discussed in the Skara
> > >> presentation from the OpenJDK Committers’ Workshop, equivalent to a
> > recent
> > >> snapshot of JDK 12 is
> > >> available for browsing and cloning from:
> > >>
> > >>     git clone https://github.com/Project-Skara/jdk.git
> > >>     https://github.com/Project-Skara/jdk
> > >>
> > >
> > > Nice! Some performance data from Germany:
> > >
> > > $ time git clone https://github.com/Project-Skara/jdk.git
> > > Cloning into 'jdk'...
> > > remote: Counting objects: 881253, done.
> > > remote: Compressing objects: 100% (160363/160363), done.
> > > remote: Total 881253 (delta 665729), reused 881253 (delta 665729),
> > pack-reused 0
> > > Receiving objects: 100% (881253/881253), 258.61 MiB | 5.32 MiB/s, done.
> > > Resolving deltas: 100% (665729/665729), done.
> > > Checking out files: 100% (66258/66258), done.
> > >
> > > real  1m25.508s
> > > user  1m8.522s
> > > sys   0m6.595s
> > >
> > > For the comparison, pulling the .hg workspace snapshot:
> > >
> > > $ time (curl https://builds.shipilev.net/workspaces/jdk-jdk.tar.xz |
> > tar xJf -;
> > > cd jdk-jdk; hg pull;
> > > hg up;)
> > >  % Total    % Received % Xferd  Average Speed   Time    Time     Time
> > Current
> > >                                 Dload  Upload   Total   Spent    Left
> > Speed
> > > 100  752M  100  752M    0     0  5673k      0  0:02:15  0:02:15
> --:--:--
> > 5737k
> > > pulling from http://hg.openjdk.java.net/jdk/jdk
> > > searching for changes
> > > adding changesets
> > > adding manifests
> > > adding file changes
> > > added 116 changesets with 1065 changes to 858 files
> > > new changesets cf34c71ca27c:802f7e5e7e6b
> > > (run 'hg update' to get a working copy)
> > > 66288 files updated, 0 files merged, 0 files removed, 0 files
> unresolved
> > >
> > >
> > > real  2m31.941s
> > > user  1m35.735s
> > > sys   0m22.859s
> > >
> > > It looks that the limiting factor here is the actual network bandwidth.
> > So, the
> > > performance is in
> > > the same ballpark, and if .hg would be smaller (my experiments saw 1.3x
> > > decreases with enabling some
> > > Mercurial options), we can get this down to 2 minutes.
> > >
> > > -Aleksey
> >
>
-- 
Cheers, Martijn (Sent from Gmail Mobile)