Sam Roberts | 9 Apr 03:09 2006

Any thoughts on BEEP over MPI?


I'm trying to figure out if MPI is a general enough message passing
protocol that you could implement BEEP over it, and whether it would
be a good idea. In particular, whether MPI could be faster than TCP.

Any pointers or thoughts would be greatly appreciated.

Thanks,
Sam
Marshall Rose | 9 Apr 18:10 2006
Picon
Picon

Re: Any thoughts on BEEP over MPI?

> I'm trying to figure out if MPI is a general enough message passing
> protocol that you could implement BEEP over it, and whether it would
> be a good idea. In particular, whether MPI could be faster than TCP.
>
> Any pointers or thoughts would be greatly appreciated.

MPI = http://www-unix.mcs.anl.gov/mpi/standard.html ??? or something  
else?

/mtr
Sam Roberts | 9 Apr 20:02 2006

Re: Any thoughts on BEEP over MPI?

Quoting mrose <at> dbc.mtview.ca.us, on Sun, Apr 09, 2006 at 09:10:41AM -0700:
> >I'm trying to figure out if MPI is a general enough message passing
> >protocol that you could implement BEEP over it, and whether it would
> >be a good idea. In particular, whether MPI could be faster than TCP.
> >
> >Any pointers or thoughts would be greatly appreciated.
> 
> MPI = http://www-unix.mcs.anl.gov/mpi/standard.html ??? or something  
> else?

Sorry, yes, that MPI.

Sam
Vijay Saraswat | 9 Apr 20:24 2006

Re: Any thoughts on BEEP over MPI?

Marshall Rose wrote:

>> I'm trying to figure out if MPI is a general enough message passing
>> protocol that you could implement BEEP over it, and whether it would
>> be a good idea. In particular, whether MPI could be faster than TCP.
>>
>> Any pointers or thoughts would be greatly appreciated.
>
>
> MPI = http://www-unix.mcs.anl.gov/mpi/standard.html ??? or something  
> else?
>
Very odd thought.

MPI is used widely as a message-passing API (particularly w/ C,C++ and 
Fortran) for high performance computing HPC). The synchronous (2-way) 
MPI is used much more widely than the 1-way (remote memory read/write) 
enabled version. It supports user-space synchronous message-passing. 
(There is also support, using immediate/nonblocking mode for 
asynchronous messaging.) The bulk of MPI is oriented towards supporting 
collective operations (e.g. all reduce) over a group of such processes 
participating in point to point interactions with each other (grouped 
together in a communicator). Not sure this is particularly relevant for 
BEEP.

Why would you want to do this? Perhaps what you might be after is that 
often HPC clusters are built over extremely high-performance 
interconnects that are non-IP-based. For instance, ARMCI and LAPI, based 
on infiniband/quadrics/myrinet/IBM Federation switches  and other 
network I/O cards
(Continue reading)

Sam Roberts | 26 Apr 07:59 2006

Re: Any thoughts on BEEP over MPI?

Quoting vijay <at> saraswat.org, on Sun, Apr 09, 2006 at 02:24:59PM -0400:
> Marshall Rose wrote:
> 
> >>I'm trying to figure out if MPI is a general enough message passing
> >>protocol that you could implement BEEP over it, and whether it would
> >>be a good idea. In particular, whether MPI could be faster than TCP.
> >>
> >>Any pointers or thoughts would be greatly appreciated.
> >
> >
> >MPI = http://www-unix.mcs.anl.gov/mpi/standard.html ??? or something  
> >else?
> >
> Very odd thought.
> 
> MPI is used widely as a message-passing API (particularly w/ C,C++ and 
> Fortran) for high performance computing HPC). The synchronous (2-way) 
> MPI is used much more widely than the 1-way (remote memory read/write) 
> enabled version. It supports user-space synchronous message-passing. 
> (There is also support, using immediate/nonblocking mode for 
> asynchronous messaging.) The bulk of MPI is oriented towards supporting 
> collective operations (e.g. all reduce) over a group of such processes 
> participating in point to point interactions with each other (grouped 
> together in a communicator). Not sure this is particularly relevant for 
> BEEP.

> Why would you want to do this? Perhaps what you might be after is that 
> often HPC clusters are built over extremely high-performance 
> interconnects that are non-IP-based. For instance, ARMCI and LAPI, based 
> on infiniband/quadrics/myrinet/IBM Federation switches  and other 
(Continue reading)

Francis Brosnan Blazquez | 28 Apr 17:00 2006
Picon

[ANN] Vortex Library 0.8.11 is ready!

       ###########################################
       ## Vortex Library 0.8.11 Stable Release  ##
       ###########################################

Release note
~~~~~~~~~~~~

  Advanced Software Production Line is proud to announce the Vortex
  Library 0.8.11 availability: an open source "BEEP Core" protocol
  implementation written in C, mapped onto TCP/IP.

  Vortex Library has been developed inside the Af-Arch project as its
  data transport layer.

  Vortex Library is a stable and robust application protocol
  development toolkit. Currently it is running under Microsoft Windows
  and GNU/Linux platforms.

  Check out the following links to get ready with Vortex Library:

    - http://vortex.aspl.es 
      [Vortex Library homepage]

    - http://www.sf.net/projects/vortexlibrary
      [Sources and Win32 binaries]

    - http://fact.aspl.es
      [Af-Arch homepage where Vortex Library is being developed]

    - http://www.beepcore.org
(Continue reading)

Marshall Rose | 28 Apr 17:18 2006
Picon
Picon

Re: [ANN] Vortex Library 0.8.11 is ready!

>   Advanced Software Production Line is proud to announce the Vortex
>   Library 0.8.11 availability: an open source "BEEP Core" protocol
>   implementation written in C, mapped onto TCP/IP.

congrats!

/mtr
Francis Brosnan Blazquez | 28 Apr 18:08 2006
Picon

Re: [ANN] Vortex Library 0.8.11 is ready!

El vie, 28-04-2006 a las 17:18 +0200, Marshall Rose escribió:
> congrats!

Thanks Marshall!!

> 
> /mtr
> 
--

-- 
Francis Brosnan Blazquez <francis <at> aspl.es>
Advanced Software Production Line, S.L.

Gmane