Le jeudi 11 décembre 2014 à 21:14 -0800, Iain Dunning a écrit :
> I'm imagine its something like the following pattern:
> 
> 
> Run 1: generate X garbage
> Run 2: generate X garbage, for total 2X garbage, which is over
> threshold, reduce back to 0
> Run 3: generate X garbage
> Run 4: generate X garbage, for total 2X garbage, which is over
> threshold, reduce back to 0
> and so on
I'm no expert on this field either, but this is probably a pattern that
can be made to give more consistent timings by the generational gc work:
https://github.com/JuliaLang/julia/pull/8699

If you're building from git master you can easily build this branch and
test it if you want. They're precisely looking for benchmarks.


Regards

> On Friday, December 12, 2014 12:09:19 AM UTC-5, Sean McBane wrote:
>         Alright. I am curious now as to what causes this behavior;
>         hopefully someone will offer an explanation.
>         
>         I'll be sure to from now on.
>         
>         -- Sean
>         
>         On Thursday, December 11, 2014 11:07:07 PM UTC-6, John Myles
>         White wrote:
>                 This is just how the GC works. Someone who's done more
>                 work on the GC can give you more context about why the
>                 GC runs for the length of time it runs for at each
>                 specific moment that it starts going. 
>                 
>                 As a favor to me, can you please make sure that you
>                 quote the entire e-mail thread you're responding to? I
>                 find responding to e-mails without context to be
>                 pretty jarring. 
>                 
>                  -- John 
>                 
>                 On Dec 12, 2014, at 12:04 AM, Sean McBane
>                 <[email protected]> wrote: 
>                 
>                 > Right, I know I'm allocating it and discarding
>                 memory. However, if the GC cleans up at deterministic
>                 points in time, as you point out in your first reply,
>                 why is timing erratic? And why the regular pattern in
>                 timing? It's always faster one call, slower one call,
>                 faster one call, slower one call... 
>                 

Reply via email to