9fans - fans of the OS Plan 9 from Bell Labs
 help / color / mirror / Atom feed
* [9fans] plan9 for calculations
@ 2009-03-27 21:31 Rudolf Sykora
  2009-03-27 23:11 ` Fernan Bolando
  2009-03-28  2:05 ` Roman V. Shaposhnik
  0 siblings, 2 replies; 3+ messages in thread
From: Rudolf Sykora @ 2009-03-27 21:31 UTC (permalink / raw)
  To: Fans of the OS Plan 9 from Bell Labs

Hello everybody,

I noticed there are some thoughts about using plan9 on supercomputers.
For me supercomputers are usually used to do some heavy calculations.
And this leads me to a question. What software is then used for
programming these calculations? (I mean e.g. linear algebra, i.e.
matrix, calculations.) Where can one read about that?

More, it also leads me to a (perhaps) simpler question. What is the
situations with ordinary machines?

Untill now I have used several libraries in linux, all of them somehow
based on lapack. I used C language (c-lapack), python (numpy), and now
I do some programming in Fortran (Intel MKL). From my experience I
would say: writing programs in C is a nightmare (for me next to no-go
again), using python with numpy is a breeze, using Fortran (95) is
sort of fine. C and Fortran run faster than python, but the factor,
when I played with it, surprised me to be sth. like 3x (expecting a
worse result).

Now I've been thinking, If I were to write sth. in plan9, what would
be the way to try?
Recently I heard about eigen2 library, which seems to be nice (high
performance, few dependencies), but for C++...

Thank you for any suggestion
Ruda

PS.: It could be that plan9, being more a os-research system than
anything else, is simply no suitable for such a use (there are no
plotting libraries, other utilities). Perhaps it's not a good idea at
all to try to use plan9 like that because it would be more work than
anything. Maybe using linux for such things with all its tools is just
ok. If you share this idea, just say it too, please.



^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [9fans] plan9 for calculations
  2009-03-27 21:31 [9fans] plan9 for calculations Rudolf Sykora
@ 2009-03-27 23:11 ` Fernan Bolando
  2009-03-28  2:05 ` Roman V. Shaposhnik
  1 sibling, 0 replies; 3+ messages in thread
From: Fernan Bolando @ 2009-03-27 23:11 UTC (permalink / raw)
  To: Fans of the OS Plan 9 from Bell Labs

On Sat, Mar 28, 2009 at 5:31 AM, Rudolf Sykora <rudolf.sykora@gmail.com> wrote:
> Hello everybody,
>
> I noticed there are some thoughts about using plan9 on supercomputers.
> For me supercomputers are usually used to do some heavy calculations.
> And this leads me to a question. What software is then used for
> programming these calculations? (I mean e.g. linear algebra, i.e.
> matrix, calculations.) Where can one read about that?
>
> More, it also leads me to a (perhaps) simpler question. What is the
> situations with ordinary machines?
>
> Untill now I have used several libraries in linux, all of them somehow
> based on lapack. I used C language (c-lapack), python (numpy), and now
> I do some programming in Fortran (Intel MKL). From my experience I
> would say: writing programs in C is a nightmare (for me next to no-go
> again), using python with numpy is a breeze, using Fortran (95) is
> sort of fine. C and Fortran run faster than python, but the factor,
> when I played with it, surprised me to be sth. like 3x (expecting a
> worse result).
>
> Now I've been thinking, If I were to write sth. in plan9, what would
> be the way to try?
> Recently I heard about eigen2 library, which seems to be nice (high
> performance, few dependencies), but for C++...
>
> Thank you for any suggestion
> Ruda
>
> PS.: It could be that plan9, being more a os-research system than
> anything else, is simply no suitable for such a use (there are no
> plotting libraries, other utilities). Perhaps it's not a good idea at
> all to try to use plan9 like that because it would be more work than
> anything. Maybe using linux for such things with all its tools is just
> ok. If you share this idea, just say it too, please.
>
>

Its not supercomputer level, but I have a sparse matrix solver in my
contrib. I use it along with haskell as simple replacement for octave.
It is mostly a collection of scripts right I am hoping to consolidate
it later as a single package. I also cannot give you a alot of example
mainly because it will translate to an actual project I did for my day
job although those were done on mathcad

I did notice that for some plotting needs simply piping it plot
command is adequate instead of bloating my tools with plot routines.

fernan

--
http://www.fernski.com



^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [9fans] plan9 for calculations
  2009-03-27 21:31 [9fans] plan9 for calculations Rudolf Sykora
  2009-03-27 23:11 ` Fernan Bolando
@ 2009-03-28  2:05 ` Roman V. Shaposhnik
  1 sibling, 0 replies; 3+ messages in thread
From: Roman V. Shaposhnik @ 2009-03-28  2:05 UTC (permalink / raw)
  To: Fans of the OS Plan 9 from Bell Labs

On 03/27/09 14:31, Rudolf Sykora wrote:
> Hello everybody,
>
> I noticed there are some thoughts about using plan9 on supercomputers.
> For me supercomputers are usually used to do some heavy calculations.
> And this leads me to a question. What software is then used for
> programming these calculations? (I mean e.g. linear algebra, i.e.
> matrix, calculations.) Where can one read about that?
>
If you are talking about established practices then supercomputing *almost*
always means Infiniband+MPI (managed by a grid engine).
On fat nodes (the kind of machines Sun used to sell) you might also find
 OpenMP. But there will always be MPI, since doesn't mean how fat the
node is -- the cluster is fatter. On the language side I've seen
predominantly
Fortran and C++, although Ron was telling me horror stories about Python
 and some other goo. As far as the libraries go: linpack is almost
always there, but the good place to take a look at what's relevant is here:
   http://docs.sun.com/app/docs/doc/819-0497
> PS.: It could be that plan9, being more a os-research system than
> anything else, is simply no suitable for such a use (there are no
> plotting libraries, other utilities). Perhaps it's not a good idea at
> all to try to use plan9 like that because it would be more work than
> anything. Maybe using linux for such things with all its tools is just
> ok. If you share this idea, just say it too, please.
>
My personal take is that Plan9's forte might be the sane clustering of
things.
MPI and grid engines is really quite horrible, but it is pervasive.

Thanks,
Roman.




^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2009-03-28  2:05 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2009-03-27 21:31 [9fans] plan9 for calculations Rudolf Sykora
2009-03-27 23:11 ` Fernan Bolando
2009-03-28  2:05 ` Roman V. Shaposhnik

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).