[Stackless] Reason for Speed Differences? Re: Ann: Stackless Psyco

Andrew Francis andrewfr_ice at yahoo.com
Sat Apr 24 21:20:10 CEST 2010


Message: 2
Date: Sat, 17 Apr 2010 15:22:14 +0000
From: Christian Tismer <tismer at stackless.com>
To: The Stackless Python Mailing List <stackless at stackless.com>
Cc: Nicholas Pi?l <nicholas at nichol.as>
Subject: Re: [Stackless] Ann: Stackless Psyco
Message-ID: <4BC9D226.8050601 at stackless.com>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed

> - Andrew Dalke's Ring Benchmark (100.000 tasklets)
>    http://www.dalkescientific.com/writings/diary/archive/2009/11/15/100000_tasklets.html
> The results are measured with 'time<bench>':
> Nbody         =  44,910sec
> Nbody w/Psyco =   9,184sec
> Ring          =   0,596sec
> Ring  w/Psyco =   1,524sec


>The Ring example goes more to GO speed, which is bad :-)

When I have time, I want to read up on strace and analyse Go. I read Andrew Dalke's blog post. From looking at Go's code, it seems to do a lot of locking and copying. Perhaps this would have an impact?



More information about the Stackless mailing list