Services & ResourcesWolfram ForumsMathGroup Archive

[Date Index] [Thread Index] [Author Index]

Re: Re: Multi-core Support in Mathematica 5.2

• To: mathgroup at smc.vnet.net
• Subject: [mg74238] Re: [mg74195] Re: Multi-core Support in Mathematica 5.2
• From: "Chris Chiasson" <chris at chiasson.name>
• Date: Thu, 15 Mar 2007 05:01:14 -0500 (EST)
• References: <esu68d$rn9$1@smc.vnet.net> <200703130303.WAA11010@smc.vnet.net>

There is the pooch mpi tool kit for Mathematica that requires specific
instructions from the user to enable parallel execution on multiple
kernels. There is also the grid computing version of Mathematica.

I don't have experience with either of these.

On 3/14/07, Neri <nkafkafi at yahoo.com> wrote:
> Thanks, Chris. I wasn't familiar with the technical term but suspected
> the answer was something like you wrote.
>
> Still, wouldn't you say that in most everyday uses the iterations in
> the Table function are independent? Wouldn't it be nice to have the
> ability to tell Mathematica that I mean them to be independent
> (perhaps through an option or attribute?) so Mathematica knows it is
> safe to parallelize the function? Alternatively, is there a way to
> replace a simple Table expression with an expression that does the
> same thing but we'll be parallelized by 5.2? I have certain
> computations that take many minutes, and basically they just repeat
> the same calculation for independent data files. Can these be
> parallelized?
>
> Someone asked a similar question in the thread at MathGroup
> http://forums.wolfram.com/mathgroup/archive/2006/Apr/msg00565.html
> and I couldn't find any answer there too. Is it a thread limitation?
>
> Thanks,
> Neri
>
> On Mar 13, 4:44 am, "Chris Chiasson" <c... at chiasson.name> wrote:
> > it depends on the code executed on iterations 1 to 10...
> >
> > technically, any mathematica code has the ability to change non-locked functions
> >
> > this could affect evaluations of parts 11 to 20
> >
> > In[1]:=
> > f[10]:=Clear@f
> > f[_]=1
> > Table[f@i,{i,1,20}]
> >
> > Out[2]=
> > 1
> >
> > Out[3]=
> > {1,1,1,1,1,1,1,1,1,Null,f[11],f[12],f[13],f[14],f[15],f[16],f[17],f[18],f[19],
> >   f[20]}
> >
> > On 3/12/07, Neri <nkafk... at yahoo.com> wrote:
> >
> >
> >
> > > Those functions that are parallelized in the 5.2, and supposedly run
> > > on more than one core simultaneously, do they run in a single thread?
> >
> > > Since it is demonstratively possible to parallelize these functions
> > > (in one thread or more) why isn't it possible to parallelize a
> > > function like Table? If I wrote a Table function with the index
> > > running from 1 to 20, isn't it possible to run 1 to 10 in one core and
> > > 11 to 20 in another core?
> >
> > > Thanks,
> > > Neri
> >
> > > > Here is a link with some details:
> >
> > > >http://www.wolfram.com/products/mathematica/newin52/multicore.html
> >
> > > > I suspect it is hard to parallelise ordinary code - even functional code
> > > > - because I can't imagine the kernel can run on more than one thread!
> >
> > > > David Baileyhttp://www.dbaileyconsultancy.co.uk
> >
> > --http://chris.chiasson.name/
>
>
>


-- 
http://chris.chiasson.name/

• References:
  • Re: Multi-core Support in Mathematica 5.2
    • From: "Neri" <nkafkafi@yahoo.com>