Home     |     .Net Programming    |     cSharp Home    |     Sql Server Home    |     Javascript / Client Side Development     |     Ajax Programming

Ruby on Rails Development     |     Perl Programming     |     C Programming Language     |     C++ Programming     |     IT Jobs

Python Programming Language     |     Laptop Suggestions?    |     TCL Scripting     |     Fortran Programming     |     Scheme Programming Language


 
 
Cervo Technologies
The Right Source to Outsource

MS Dynamics CRM 3.0

Python Programming Language

Can Parallel Python run on a muti-CPU server ?


Hi all,

I'm interested in Parallel Python and I learned from the website of
Parallel Python
that it can run on SMP and clusters. But can it run on a our muti-CPU
server ?
We are running an origin3800 server with 128 CPUs.

Thanks.

fdu.xia@gmail.com wrote:
> Hi all,

> I'm interested in Parallel Python and I learned from the website of
> Parallel Python
> that it can run on SMP and clusters. But can it run on a our muti-CPU
> server ?
> We are running an origin3800 server with 128 CPUs.

> Thanks.

I have tested that at least it could run sum_primes.py on our server.

But it seems Parallel Python just launch one python process for
each job, and if I let it use 12 CPUs for 8 jobs, Parallel Python
launches 12 python processes, 4 of which just sleep until all 8 jobs
are done.

>From the www.parallelpython.com , the 'Features' section:

--------------------------------
Features:
 *Parallel execution of python code on SMP and clusters
-------------------------------

PP uses processes, and thus it will take advantage of multiple cores
for a CPU bound task.

-Nick

On Feb 6, 9:13 pm, "fdu.xia@gmail.com" <fdu.xia@gmail.com>
wrote:

I've just downloaded it ,having a couple of M-CPU machines, it's quite
interesting. At this moment I think that you should not see it as
'magical distribution' of your processing pool but more like the way
threads work. So every function will stay on it's CPU while executing
and will not use the process power of another CPU if available.

So I guess you have to fine grain your program to take the advantage of
multiple CPU's, just like you would do if you had 'real' threads.

--
mph

On Feb 7, 3:13 am, "fdu.xia@gmail.com" <fdu.xia@gmail.com>
wrote:

> Hi all,

> I'm interested in Parallel Python and I learned from the website of
> Parallel Python
> that it can run on SMP and clusters. But can it run on a our muti-CPU
> server ?
> We are running an origin3800 server with 128 CPUs.

> Thanks.

I see you got a problem. me to. how can i get such a little toy for my
own and how much do i have to spend on it.
i also want sometjing like thi. 128 cpu-s, 64 GB rainbowtables, oh my
god. i want this. i am very close to have an orgasm.

Rent it :-) Well you could do at Sara, just google for "sara teras" ;-)

--
mph

no, not renting. i need such one at home. when you say rent it sounds
like buy a hosting package. i need one to work all the time on max
power. cracking md5 needs power. :-D
Hi,

That is definitely possible!
To achieve the best performance split your calculation either into 128
equal parts or int >>128 part of any size (then load balancing will
spread workload equally). Let us know the results, if need any help
with parallelization feel free to request it here:
http://www.parallelpython.com/component/option,com_smf/Itemid,29/
Thank you!

On Feb 7, 2:13 am, "fdu.xia@gmail.com" <fdu.xia@gmail.com>
wrote:

Add to del.icio.us | Digg this | Stumble it | Powered by Megasolutions Inc