[Pharo-dev] Pharo vs. Squeak performance

Clément Bera bera.clement at gmail.com
Tue Jun 21 10:46:37 EDT 2016


On Tue, Jun 21, 2016 at 3:13 PM, Sven Van Caekenberghe <sven at stfx.eu> wrote:

>
> > On 21 Jun 2016, at 11:58, Ben Coman <btc at openinworld.com> wrote:
> >
> > On Tue, Jun 21, 2016 at 3:29 PM, Clément Bera <bera.clement at gmail.com>
> wrote:
> >> Hello John.
> >>
> >> I'm just guessing here. Lacking information. It could be:
> >>
> >> Guess 3) the UI is known to be much slower in Pharo. Can you try
> headless or
> >> after ticking "Server mode" In the Pharo settings in System.
> >
> > Or try something like...
> > [ Transcript cr; show: (Time millisecondsToRun: [1 to: 100000000 do:
> > [:i | Object new]] ) ] forkAt: 75.
>
> BTW, this is essentially a garbage collection benchmark: you create
> 100,000,000 empty objects. It stresses the GC, especially the ephemeral
> phase of it. Since the GC is part of the VM, you are testing the VM more
> than any image code.
>

Right, it could be related to the new finalization ...

We can't guess properly we have no VM versions.

>
> > cheers -ben
> >
> >>
> >>
> >>
> >>
> >>
> >> On Tue, Jun 21, 2016 at 1:28 AM, John Brant <brant at refactoryworkers.com
> >
> >> wrote:
> >>>
> >>> I have some code that creates a several hundred MB model. When I run
> the
> >>> code under Pharo it takes ~2.5 minutes to run. However, if I run the
> same
> >>> code in Squeak, it takes ~2 minutes. Since my code just uses base
> >>> collections and streams, I thought the times should be very similar
> between
> >>> the two. After a little investigation, I noticed that even simple
> things
> >>> like “Object new” can take much more time in Pharo. Here’s an example
> that I
> >>> executed in Squeak and Pharo:
> >>>
> >>>        Time millisecondsToRun: [1 to: 100000000 do: [:i | Object new]]
> >>>
> >>>        Squeak times:
> >>>                1255 1257 1261 1265 1280 1294 1314 1337 1350 1360
> >>>
> >>>        Pharo times:
> >>>                1815 1818 1870 1879 1900 1922 1944 1952 1958 2170
> >>>
> >>> The results are the first 10 executions sorted by time after opening an
> >>> image. Pharo doesn’t always give these poor results. Occasionally I
> can get
> >>> times as good a Squeak. For example, I was able to get these times in
> Pharo:
> >>> 1253, 1284, 1297, 1314, 1317. However, it generally takes ~1.8 seconds
> in
> >>> Pharo vs. the ~1.3 seconds for Squeak. The worst time I got for Squeak
> was
> >>> in the 1.6 second range. The worst for Pharo was in the 4.3 second
> range.
> >>>
> >>> Does anyone know why Pharo is slower? Is there some memory setting
> that I
> >>> need to change?
> >>>
> >>>
> >>> John Brant
> >>
> >>
> >
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.pharo.org/pipermail/pharo-dev_lists.pharo.org/attachments/20160621/5fad62ce/attachment-0002.html>


More information about the Pharo-dev mailing list