1
1
openmpi/README.alpha
Jeff Squyres 7d9ea5b084 Add note to packagers
This commit was SVN r5144.
2005-04-02 13:59:08 +00:00

389 строки
14 KiB
Plaintext

Copyright (c) 2004-2005 The Trustees of Indiana University.
All rights reserved.
Copyright (c) 2004-2005 The Trustees of the University of Tennessee.
All rights reserved.
Copyright (c) 2004-2005 High Performance Computing Center Stuttgart,
University of Stuttgart. All rights reserved.
Copyright (c) 2004-2005 The Regents of the University of California.
All rights reserved.
$COPYRIGHT$
Additional copyrights may follow
$HEADER$
*** NOTE TO PACKAGERS: Do *not* include this file in the package.
This file will only be included in the alpha release. Proper
documentation will be provided instead of this file.
Greetings!
The Open MPI development team is pleased to announce an alpha-quality
release of the current state of our software to a limited set of
"friends" -- those who are knowledgeable about MPI and can provide
intelligent feedback to us before a public release. Our intent with
this release is to get the Open MPI software onto other people's
machines and see what creative ways you can come up with to break it,
and to send us your comments and suggestions.
However, we want to stress the following points:
- This is an *alpha* quality release. We're quite aware that there
are several things still broken (but report them anyway!).
- Most notably, if you run basic performance testing, you'll notice
that, for example, the GM numbers are still a microsecond or two too
high (we've been concentrating on functionality for the last month
-- performance tuning is coming shortly)
- Since the competition in the HPC community is rather fierce, please
do not redistribute this software without our permission. Also,
please do not publish any results (good or bad) because, as
mentioned above, this is pre-release software and we still have
performance tuning to do.
===========================================================================
You have probably downloaded this tarball from
http://www.open-mpi.org/nightly/. This directory is updated at least
once a day around 2am US/Indiana time (assuming that there is new code
to release). It may be updated more frequently if a critical bug fix
is reported and fixed.
The best way to report bugs, send comments, or ask questions is to
sign up on the devel@open-mpi.org mailing list:
http://www.open-mpi.org/mailman/listinfo.cgi/devel
Thanks for your time.
===========================================================================
This tarball is an alpha release of Open MPI. It is not yet complete,
mainly in the following areas:
- Support for Infiniband (both verbs and OpenIB) is missing
- Support for Quadrics is missing
- Support for Myrinet needs performance tuning
- Support for MX needs performance tuning
- Support for TCP needs performance tuning
- Support for shared memory is not fully debugged
--> If this becomes a problem during your testing, run the
following:
shell$ rm -f <prefix>/lib/openmpi/*sm*
where <prefix> is the directory where you installed Open MPI.
- Striping MPI messages across multiple networks is supported (and
happens automatically when multiple networks are available), but
needs performance tuning
- The only run-time systems supported are:
- rsh / ssh
- BProc of the flavor that is used at Los Alamos National Labs (in
particular, it must be used with the BJS scheduler)
- Complete user and system administrator documentation is missing
(this file comprises the majority of the current user documentation)
- The only systems that have been tested on are:
- Linux, 32 bit, with gcc
- Other systems have been lightly (but not fully tested):
- Linux, 64 bit, with gcc
- OS X (10.3), 32 bit, with gcc
- Missing MPI functionality:
- The Fortran 90 MPI API is disabled (it is not complete).
- MPI-2 dynamic functionality is temporarily broken.
- MPI-2 one-sided functionality will not be included in the first
few releases of Open MPI
- After running MPI applications, the directory
/tmp/openmpi-sessions-<username>@<hostname>* will exist (but will
likely be empty). It is safe to remove.
===========================================================================
Building Open MPI
-----------------
Open MPI uses a traditional configure script paired with "make" to
build. Typical installs can be of the pattern:
---------------------------------------------------------------------------
shell$ ./configure [...options...]
shell$ make all install
---------------------------------------------------------------------------
There are many available configure options; a summary of the more
important ones follows:
--prefix=<directory>
Install Open MPI into the base directory named <directory>. Hence,
Open MPI will place its executables in <directory>/bin, its header
files in <directory>/include, its libraries in <directory>/lib, etc.
--with-ptl-gm=<directory>
Specify the directory where the GM libraries and header files are
located. This enables GM support in Open MPI.
--with-mpi-param_check(=value)
"value" can be one of: always, never, runtime. If no value is
specified, or this option is not used, "always" is the default.
Using --without-mpi-param-check is equivalent to "never".
- always: the parameters of MPI functions are always checked for
errors
- never: the parameters of MPI functions are never checked for
errors
- runtime: whether the parameters of MPI functions are checked
depends on the value of the MCA parameter mpi_param_check
(default: yes).
--with-threads=value
Since thread support (both support for MPI_THREAD_MULTIPLE and
asynchronous progress) is only partially tested, it is disabled by
default. To enable threading, use "--with-threads=posix". This is
most useful when combined with --enable-mpi-threads and/or
--enable-progress-threads.
--enable-mpi-threads
Allows the MPI thread level MPI_THREAD_MULTIPLE. See
--with-threads; this is currently disabled by default.
--enable-progress-threads
Allows asynchronous progress in some transports. See
--with-threads; this is currently disabled by default.
--disable-shared
By default, libmpi is built as a shared library, and all components
are built as dynamic shared objects (DSOs). This switch disables
this default; it is really only useful when used with
--enable-static.
--enable-static
Build libmpi as a static library, and statically link in all
components.
There are other options available -- see "./configure --help".
Open MPI supports all the "make" targets that are provided by GNU
Automake, such as:
all - build the entire package
install - install the package
uninstall - remove all traces of the package from the $prefix
clean - clean out the build tree
Once Open MPI has been built and installed, it is safe to run "make
clean" and/or remove the entire build tree.
VPATH builds are supported.
Generally speaking, the only thing that users need to do to use Open
MPI is ensure that <prefix>/bin is in their PATH. Users may need to
ensure that this directory is set in their PATH in their shell setup
files (e.g., .bashrc, .cshrc) so that rsh/ssh-based logins will be
able to find the Open MPI executables.
Setting LD_LIBRARY_PATH is typically not necessary, but in some cases,
if libmpi.so cannot be found when MPI applications are run,
LD_LIBRARY_PATH can be set to <prefix>/lib.
===========================================================================
Checking Your Open MPI Installation
-----------------------------------
The "ompi_info" command can be used to check the status of your Open
MPI installation (located in <prefix>/bin/ompi_info). Running it with
no arguments provides a summary of information about your Open MPI
installation.
Note that the ompi_info command is extremely helpful in determining
which components are installed as well as listing all the run-time
settable parameters that are available in each component (as well as
their default values).
The following options may be helpful:
--all Show a *lot* of information about your Open MPI
installation.
--parsable Display all the information in an easily
grep/cut/awk/sed-able format.
--param <framework> <component>
A <framework> of "all" and a <component> of "all" will
show all parameters to all components. Otherwise, the
parameters of all the components in a specific framework,
or just the parameters of a specific component can be
displayed by using an appropriate <framework> and/or
<component> name.
Changing the values of these parameters is explained in the "The
Modular Component Architecture (MCA)" section, below.
===========================================================================
Compiling Open MPI Applications
-------------------------------
Open MPI provides "wrapper" compilers that should be used for
compiling MPI applications:
mpicc
mpiCC (or mpiC++ if your filesystem is case-insensitive)
mpif77
mpif90
For example:
shell$ mpicc hello_world_mpi.c -o hello_world_mpi -g
shell$
All the wrapper compilers do is add a variety of compiler and linker
flags to the command line and then invoke a back-end compiler. The
end result is an MPI executable that is properly linked to all the
relevant libraries.
===========================================================================
Running Open MPI Applications
-----------------------------
Open MPI supports both mpirun and mpiexec (they are actually the
same). For example:
shell$ mpirun -np 2 hello_world_mpi
or
shell$ mpiexec -np 1 hello_world_mpi : -np 1 hello_world_mpi
are equivalent. Many of mpiexec's switches (such as -host and -arch)
are not yet functional, although they will not error if you try to use
them.
Since rsh is probably the launcher that you will be using (if you are
outside of Los Alamos National Laboratory), you can also specify a
-hostfile or -machinefile parameter, indicating an standard
mpirun-style hostfile (one hostname per line):
shell$ mpirun -hostfile my_hostfile -np 2 hello_world_mpi
If you intend to run more than one process on a node, the hostfile can
use the "cpu" attribute. If "cpu" is not specified, a count of 1 is
assumed. For example, using the following hostfile:
---------------------------------------------------------------------------
node1.example.com
node2.example.com
node3.example.com cpu=2
node4.example.com cpu=4
---------------------------------------------------------------------------
shell$ mpirun -hostfile my_hostfile -np 8 hello_world_mpi
will launch MPI_COMM_WORLD rank 0 on node1, rank 1 on node2, ranks 2
and 3 on node3, and ranks 4 through 7 on node4.
Note that the values of component parameters can be changed on the
mpirun / mpiexec command line. This is explained in the section
below, "The Modular Component Architecture (MCA)".
===========================================================================
The Modular Component Architecture (MCA)
The MCA is the backbone of Open MPI -- most services and functionality
are implemented through MCA components. Here is a list of all the
component frameworks in Open MPI:
---------------------------------------------------------------------------
MPI component frameworks:
-------------------------
coll - MPI collective algorithms
io - MPI-2 I/O
pml - MPI point-to-point management layer
ptl - MPI point-to-point transport layer
topo - MPI topology routines
Back-end run-time environment component frameworks:
---------------------------------------------------
errmgr - RTE error manager
gpr - General purpose registry
iof - I/O forwarding
ns - Name server
oob - Out of band messaging
pls - Process launch system
ras - Resource allocation system
rds - Resource discovery system
rmaps - Resource mapping system
rmgr - Resource manager
rml - RTE message layer
soh - State of health monitor
Miscellaneous frameworks:
-------------------------
allocator - Memory allocator
mpool - Memory pooling
---------------------------------------------------------------------------
Each framework typically has one or more components that are used at
run-time. For example, the ptl framework is used by MPI to send bytes
across underlying networks. The tcp ptl, for example, sends messages
across TCP-based networks; the gm ptl sends messages across GM
Myrinet-based networks.
Each component typically has some tunable parameters that can be
changed at run-time. Use the ompi_info command to check a component
to see what its tunable parameters are. For example:
shell$ ompi_info --param ptl tcp
shows all the parameters (and default values) for the tcp ptl
component.
These values can be overridden at run-time in several ways. At
run-time, the following locations are examined (in order) for new
values of parameters:
1. <prefix>/etc/openmpi-mca-params.conf
A simple text file with lots of comments explaining its format.
This file is intended to set any system-wide default MCA parameter
values -- it will apply, by default, to all users who use this Open
MPI installation.
2. $HOME/.openmpi/mca-params.conf
If this file exists, it should be in the same format as
<prefix>/etc/openmpi-pmca-params.conf. It is intended to provide
per-user default parameter values.
3. environment variables of the form OMPI_MCA_<name> set equal to a <value>
Where <name> is the name of the parameter. For example, set the
variable named OMPI_MCA_ptl_tcp_frag_size to the value 65536.
4. the mpirun command line: --mca <name> <value>
Where <name> is the name of the parameter. For example:
mpirun --mca ptl_tcp_frag_size 65536 -np 2 hello_world_mpi
These locations are checked in order; for example, a parameter value
passed on the mpirun command line will override an environment
variable, and an environment variable will override the system-wide
defaults.
===========================================================================
Got more questions?
-------------------
The best way to report bugs, send comments, or ask questions is to
sign up on the devel@open-mpi.org mailing list:
http://www.open-mpi.org/mailman/listinfo.cgi/devel
Thanks for your time.