IBM PE for AIX V2R4.0: Operation and Use, Vol. 1
This appendix contains the manual pages for the PE commands discussed
throughout this book. Each manual page is organized into the sections
listed below. The sections always appear in the same order, but some
appear in all manual pages while others are optional.
- NAME
- Provides the name of the command described in the manual page, and a brief
description of its purpose.
- SYNOPSIS
- Includes a diagram that summarizes the command syntax, and provides a
brief synopsis of its use and function. If you are unfamiliar with the
typographic conventions used in the syntax diagrams, see "Typographic Conventions".
- FLAGS
- Lists and describes any required and optional flags for the
command.
- DESCRIPTION
- Describes the command more fully than the NAME and
SYNOPSIS sections.
- ENVIRONMENT VARIABLES
- Lists and describes any applicable environment variables.
- EXAMPLES
- Provides examples of ways in which the command is typically used.
- FILES
- Lists and describes any files related to the command.
- RELATED INFORMATION
- Lists commands, functions, file formats, and special files that are
employed by the command, that have a purpose related to the command, or that
are otherwise of interest within the context of the command.
NAME
mcp - Allows you to propagate a copy of a file to multiple
nodes on an IBM POWERparallel system.
SYNOPSIS
mcp infile [outfile] [POE options]
In the command synopsis above, the infile is the name of the file to be
copied. You can copy to a new name by specifying an outfile. If
you do not provide the outfile name, the file will be placed in its current
directory on each node. The outfile can be either an explicit output
file name or a directory name. When a directory is specified, the file
is copied with the same name to that directory.
DESCRIPTION
The mcp command allows you to propagate a copy of a file to
multiple nodes on an IBM RS/6000 SP. The file must initially reside (or
be NFS-mounted) on at least one node.
mcp is a POE program and, therefore, all POE options are
available. You can set POE options with either command line flags or
environment variables. The number of nodes to copy the file to
(-procs), and the message passing protocol used to copy the file
(-euilib) are the POE options of most interest. The input
file must be readable from the node assigned to task 0.
| Note: | A POE job loads faster if a copy of the job resides on each node. For
this reason, it is suggested that you use mcp to copy your
executable to a file system such as /tmp, which resides on each node.
|
Return codes are:
| Note: | The actual command return code value is 128 plus the unsigned return code
value. That is, a return code of -2 will give a value of 130.
For more information, see the "Exit Status" section in IBM Parallel
Environment for AIX: MPI Programming and Subroutine Reference
|
- -1
- incorrect usage
- -2
- error opening input file
- -3
- error opening to file on originating node
- -4
- error writing data to to file on originating node
- -5
- no room on remote node's file system
- -6
- error opening file on remote node
- -7
- error writing data on remote node
- -8
- error renaming temp file to file name
- -9
- input file is empty
- -10
- invalid block size
- -11
- error allocating storage
ENVIRONMENT VARIABLES
- MP_BLKSIZE
- sets the block size used for copying the data. This can be a value
between 1 and 8,000,000 (8 megabytes). The default is 100,000
(100K).
EXAMPLES
- To copy a file from your current directory to the current directory on all
nodes of a 16-processor system, using the High Performance Switch,
enter:
mcp filename -procs 16 -euilib us
- To copy a filename from your current directory to the /tmp directory on
all nodes of a 16-processor system, using IP, enter:
mcp filename /tmp -procs 16 -euilib ip
- To copy a file from your current directory to a different filename on all
nodes of a 16-processor system, enter:
mcp filename /tmp/newfilename -procs 16
RELATED INFORMATION
Commands: mprcp(1), rcp(1)
NAME
mcpgath - Takes files from each task of tasks 0 through task
N and copies them back in sequence to task 0.
SYNOPSIS
mcpgath [-ai] source ... destination [POE options]
Source is one of the following:
- one or more existing file names - files will be copied with the same names
to the destination directory on task 0. Each file name specified must
exist on all tasks involved in the copy.
- a directory name - all files in that directory on each task are copied
with the same names to the destination directory on task 0.
- an expansion of file names, using wildcards - files are copied with the
same names to the destination directory. All wildcarded input strings
must be enclosed in double quotes.
Destination is an existing destination directory name to where the data
will be copied. The destination directory must be the last item
specified before any POE flags.
FLAGS
- -a
- An optional flag that appends the task number to the end of the file name
when it is copied to task 0. This is for task identification purposes,
to know where the data came from. The -a and -i
flags can be combined to check for existing files appended with the task
number.
- -i
- An optional flag that checks for duplicate or existing files of the same
name, and does not replace any existing file found. Instead, issues an
error message and continues with the remaining files to be copied. The
-a and -i flags can be combined to check for existing
files appended with the task number.
See "Executing Parallel Programs" for information on POE options.
DESCRIPTION
The mcpgath function determines the list of files to be gathered
on each task. This function also resolves the source file, destination
directory, and path names with any meta characters, wildcard expansions,
etc. to come up with valid file names. Wildcards should be
enclosed in double quotes, otherwise they will be expanded locally on the task
from where the command is issued, which may not produce the intended file name
resolution.
mcpgath is a POE program and, therefore, all POE options are
available. You can set POE options with either command line flags or
environment variables. The number of nodes to copy the file to
(-procs), and the message passing protocol used to copy the file
(-euilib) are the POE options of most interest.
| Note: | A default of 100K data block size is used for copying the data. This
can be changed by updating the source found in
/usr/lpp/ppe.poe/samples/mpi, and compiling it with the
mpcc command.
|
Return codes are:
| Note: | The actual command return code value is 128 plus the unsigned return code
value. That is, a return code of -2 will give a value of 130.
For more information, see the "Exit Status" section in IBM Parallel
Environment for AIX: MPI Programming and Subroutine Reference
|
- -1
- invalid number of arguments specified
- -2
- invalid option flag specified
- -3
- unable to resolve input file name(s)
- -4
- could not open input file for read
- -5
- no room on destination node's file system
- -6
- error opening file output file
- -7
- error creating output file
- -8
- error writing to output file
- -9
- MPI_Send of data failed
- -10
- final MPI_Send failed
- -11
- MPI_Recv failed
- -12
- invalid block size
- -13
- error allocating storage
- -14
- total number of tasks must be greater than one
ENVIRONMENT VARIABLES
- MP_BLKSIZE
- sets the block size used for copying the data. This can be a value
between 1 and 8,000,000 (8 megabytes). The default is 100,000
(100K).
EXAMPLES
- You can copy a single file from all tasks into the destination
directory. For example, enter:
mcpgath -a hello_world /tmp -procs 4
This will copy the file hello_world (assuming it is a file and
not a directory) from tasks 0 through 3 as to task 0:
From task 0: /tmp/hello_world.0
From task 1: /tmp/hello_world.1
From task 2: /tmp/hello_world.2
From task 3: /tmp/hello_world.3
- You can specify any number of files as source files. The
destination directory must be the last item specified before any POE
flags. For example:
mcpgath -a file1.a file2.a file3.a file4.a file5.a /tmp -procs 4
will take file1.a through file5.a from
the local directory on each task and copy them back to task 0. All
files specified must exist on all tasks involved. The file distribution
will be as follows:
From Task 0: /tmp/file1.a.0
From Task 1: /tmp/file1.a.1
From Task 2: /tmp/file1.a.2
From Task 3: /tmp/file1.a.3
From Task 0: /tmp/file2.a.0
From Task 1: /tmp/file2.a.1
From Task 2: /tmp/file2.a.2
From Task 3: /tmp/file2.a.3
From Task 0: /tmp/file3.a.0
From Task 1: /tmp/file3.a.1
From Task 2: /tmp/file3.a.2
From Task 3: /tmp/file3.a.3
From Task 0: /tmp/file4.a.0
From Task 1: /tmp/file4.a.1
From Task 2: /tmp/file4.a.2
From Task 3: /tmp/file4.a.3
From Task 0: /tmp/file5.a.0
From Task 1: /tmp/file5.a.1
From Task 2: /tmp/file5.a.2
From Task 3: /tmp/file5.a.3
- You can specify wildcard values to expand into a list of files to be
gathered. For this example, assume the following distribution of files
before calling mcpgath:
Task 0 contains file1.a and file2.a
Task 1 contains file1.a only
Task 2 contains file1.a, file2.a, and file3.a
Task 3 contains file4.a, file5.a, and file6.a
Enter:
mcpgath -a "file*.a" /tmp -procs 4
This will pass the wildcard expansion to each task, which will resolve
into the list of locally existing files to be copied. This will result
in the following distribution of files on task 0:
From Task 0: /tmp/file1.a.0
From Task 0: /tmp/file2.a.0
From Task 1: /tmp/file1.a.1
From Task 2: /tmp/file1.a.2
From Task 2: /tmp/file2.a.2
From Task 2: /tmp/file3.a.2
From Task 3: /tmp/file4.a.3
From Task 3: /tmp/file5.a.3
From Task 3: /tmp/file6.a.3
- You can specify a directory name as the source, from which the files to be
gathered are found. For this example, assume the following distribution
of files before calling mcpgath:
Task 0 /test contains file1.a and file2.a
Task 1 /test contains file1.a only
Task 2 /test contains file1.a and file3.a
Task 3 /test contains file2.a, file4.a, and file5.a
Enter:
mcpgath -a /test /tmp -procs 4
This results in the following file distribution:
From Task 0: /tmp/file1.a.0
From Task 0: /tmp/file2.a.0
From Task 1: /tmp/file1.a.1
From Task 2: /tmp/file1.a.2
From Task 2: /tmp/file3.a.2
From Task 3: /tmp/file2.a.3
From Task 3: /tmp/file4.a.3
From Task 3: /tmp/file5.a.3
NAME
mcpscat - Takes a number of files from task 0 and scatters
them in sequence to all tasks, in a round robin order.
SYNOPSIS
mcpscat [-f] [-i] source ... destination
[POE options]
Source can be one of the following:
- a single file name - file is copied to all tasks
- a single file name that contains a list of file names (-f
option)
- two of more file names - files will be distributed in a round robin order
to the tasks
- an expansion of file names, using wildcards - files will be distributed in
a round robin order to the tasks
- a directory name - all files in that directory are copied in a round robin
order to the tasks.
Destination is an existing destination directory name to where the data
will be copied.
FLAGS
- -f
- Is an optional flag that says the first file contains the names of the
source files that are to be scattered. Each file name, in the file,
must be specified on a separate line. No wildcards are supported when
this option is used. Directory names are not supported in the file
either. When this option is used, the mcpscat parameters
should consist of a single source file name (for the list of files) and a
destination directory. The files will then be scattered just as if they
had all been specified on the command line in the same order as they are
listed in the file.
- -i
- Checks for duplicate or existing files of the same name, and does not
replace any existing file found. Instead, issues an error message and
continues with the remaining files to be copied. Without this flag, the
default action is to replace any existing files with the source file.
See "Executing Parallel Programs" for information on POE options.
DESCRIPTION
The mcpscat function determines the order in which to distribute
the files, using a round robin method, according to the list of nodes and
number of tasks. Files are sent in a one-to-one correspondence to the
nodes in the list of tasks. If the number of files specified is greater
than the number of nodes, the remaining files are sent in another round
through the list of nodes. Wildcards should be enclosed in double
quotes, otherwise they will be expanded locally on the task from where the
command is issued, which may not produce the intended file name
resolution.
mcpscat is a POE program and, therefore, all POE options are
available. You can set POE options with either command line flags or
environment variables. The number of nodes to copy the file to
(-procs), and the message passing protocol used to copy the file
(-euilib) are the POE options of most interest.
Return codes are:
| Note: | The actual command return code value is 128 plus the unsigned return code
value. That is, a return code of -2 will give a value of 130.
For more information, see the "Exit Status" section in IBM Parallel
Environment for AIX: MPI Programming and Subroutine Reference
|
- -1
- invalid number of arguments specified
- -2
- invalid option flag specified
- -3
- unable to resolve input file name(s)
- -4
- could not open input file for read
- -5
- no room on destination node's file system
- -6
- error opening file output file
- -7
- error creating output file
- -8
- MPI_Send of data failed
- -9
- final MPI_Send failed
- -10
- MPI_Recv failed
- -11
- failed opening temporary file
- -12
- failed writing temporary file
- -13
- error renaming temp file to filename
- -14
- input file is empty (zero byte file size)
- -15
- invalid block size
- -16
- error allocating storage
- -17
- number of tasks and files do not match
- -18
- not enough memory for list of file names
ENVIRONMENT VARIABLES
- MP_BLKSIZE
- sets the block size used for copying the data. This can be a value
between 1 and 8,000,000 (8 megabytes). The default is 100,000
(100K).
EXAMPLES
- You can copy a single file to all tasks into the destination
directory. For example, enter:
mcpscat filename /tmp -procs 4
This will take the file and distribute it to tasks 0 through 3 as
/tmp/filename.
- You can specify any number of files as source files. The
destination directory must be the last item specified before any POE
flags. For example:
mcpscat file1.a file2.a file3.a file4.a file5.a /tmp -procs 4
will take file1.a through file5.a from
the local directory and copy them in a round robin order to tasks 0 through 3
into /tmp. The file distribution will be as follows:
Task 0: /tmp/file1.a
Task 1: /tmp/file2.a
Task 2: /tmp/file3.a
Task 3: /tmp/file4.a
Task 0: /tmp/file5.a
- You can specify the source files to copy in a file. For
example:
mcpscat -f file.list /tmp -procs 4
will produce the same results as the previous example if as
file.list contains five lines with the file names
file1.a through file5.a in it.
- You can specify wildcard values to expand into a list of files to be
scattered. Enter:
mcpscat "file*.a" /tmp -procs 4
Assuming Task 0 contains file1.a,
file2.a, file3.a, file4.a, and
file5.a in its home directory, this will result in a similar
distribution as in the previous example.
- You can specify a directory name as the source, from which the files to be
scattered are found. Assuming /test contains
myfile.a, myfile.b, myfile.c,
myfile.d, myfile.f, and
myfile.g on Task 0, enter:
mcpscat /test /tmp -procs 4
This results in the following file distribution:
Task 0: /tmp/myfile.a
Task 1: /tmp/myfile.b
Task 2: /tmp/myfile.c
Task 3: /tmp/myfile.d
Task 0: /tmp/myfile.f
Task 1: /tmp/myfile.g
NAME
mpamddir - echoes an amd-mountable directory name.
SYNOPSIS
mpamddir
or, if you're using the Parallel Environment for AIX:
export MP_REMOTEDIR=mpamddir
This script determines whether or not the current directory is a mounted
file system. If it is, it looks to see if it appears in the amd maps,
and constructs a name for the directory that is known to amd. You can
modify this script, or create additional ones that apply to your
installation.
By default, POE uses the Korn shell pwd command to obtain the
name of the current directory to pass to the remote nodes for
execution. This works for C shell users if the current directory
is:
- The home directory
- Not mounted by amd, the AutoMount Daemon.
If this is not the case, (for example, if the user's current directory
is a subdirectory of the home directory), then you can supply your own script
for providing the name of the current directory on the remote nodes.
To use mpamddir as the script for providing the name, export the
environment variable MP_REMOTEDIR, and set it to
mpamddir.
RELATED INFORMATION
Commands: ksh(1), poe(1), csh(1)
NAME
mpcc - Invokes a shell script to compile C programs.
SYNOPSIS
mpcc [cc_flags]... program.c
The mpcc shell script compiles C programs while linking in the
Partition Manager, Message Passing Interface (MPI), and/or Message Passing
Library (MPL).
FLAGS
Any of the compiler flags normally accepted by the cc command can
also be used on mpcc. For a complete listing of these flag
options, refer to the manual page for the AIX cc command.
Typical options to mpcc include:
- -v
- causes a "verbose" output listing of the shell script.
- -g
- Produces an object file with symbol table references. This object
file is needed for debugging with the pdbx and pedb
debuggers, and is also needed by the Source Code view of the
Visualization Tool (vt).
- -o
- names the executable.
- -l (lower-case L)
- names additional libraries to be searched. Several libraries are
automatically included, and are listed below in the FILES section.
- -I (upper-case i)
- names directories for additional includes. The directory
/usr/lpp/ppe.poe/include is automatically included.
- -p
- enables profiling with the prof command. For more
information, see the appendix on "Profiling Programs" in IBM
Parallel Environment for AIX: Operation and Use, Volume 2, Tools
Reference
- -pg
- enables profiling with the xprofiler and gprof
commands. For more information, see the "Xprofiler" chapter and
the appendix on "Profiling Programs" in IBM Parallel Environment for
AIX: Operation and Use, Volume 2, Tools Reference
DESCRIPTION
The mpcc shell script calls the AIX xlc compiler.
In addition, the Partition Manager and message passing interface are
automatically linked in. The script creates an executable that
dynamically binds with the message passing libraries. If you wish to
create a statically bound application, use the instructions in "Creating a Static Executable" in place of this script.
Flags are passed by mpcc to the xlc command, so any of
the xlc options can be used on the mpcc shell script.
The communication subsystem library implementation is dynamically linked when
you invoke the executable using the poe command. The value
specified by the MP_EUILIB environment variable or the
-euilib flag will then determine which communication subsystem
library implementation is dynamically loaded.
ENVIRONMENT VARIABLES
- MP_PREFIX
- sets an alternate path to the scripts library. If not set or NULL,
the standard path /usr/lpp/ppe.poe is used. If this
environment variable is set, then all libraries are prefixed by
$MP_PREFIX/ppe.poe.
EXAMPLES
To compile a C program, enter:
mpcc program.c -o program
FILES
When you compile a program using mpcc, the following libraries are
automatically selected:
- /usr/lpp/ppe.poe/lib/libmpi.a (Message Passing Interface,
collective communication routines)
- /usr/lpp/ppe.poe/lib/libvtd.a (VT tracing subsystem)
- /usr/lpp/ppe.poe/lib/libppe.a (PE common routines)
- /usr/lpp/ppe.poe/lib/libc.a (POE version of libc.a)
RELATED INFORMATION
Commands: mpcc_r(1), mpCC(1), mpCC_r(1),
mpxlf(1), cc(1), pdbx(1), pedb(1),
vt(1), xprofiler(1)
NAME
mpcc_chkpt - Invokes a shell script to compile
checkpointable C programs.
SYNOPSIS
mpcc_chkpt [cc_flags]... -us | -ip program.c
The mpcc_chkpt shell script compiles C programs while linking in
the Partition Manager, Message Passing Interface (MPI), and support code for
parallel Checkpoint/Restart. It builds an executable with no shared
objects.
FLAGS
Most of the compiler flags normally accepted by the cc command can
also be used on mpcc_chkpt. For a complete listing of these
flag options, refer to the manual page for the AIX cc command.
Typical options to mpcc_chkpt include:
- -v
- causes a "verbose" output listing of the shell script.
- -g
- Produces an object file with symbol table references. This object
file is needed for debugging with the pdbx and pedb
debuggers, and is also needed by the Source Code view of the
Visualization Tool (vt).
- -o
- names the executable.
- -l (lower-case L)
- names additional libraries to be searched. Several libraries are
automatically included, and are listed below in the FILES section.
- -I (upper-case i)
- names directories for additional includes. The directory
/usr/lpp/ppe.poe/include is automatically included.
- -p
- enables profiling with the prof command. For more
information, see the appendix on "Profiling Programs" in IBM
Parallel Environment for AIX: Operation and Use, Volume 2, Tools
Reference
- -pg
- enables profiling with the xprofiler and gprof
commands. For more information, see the "Xprofiler" chapter and
the appendix on "Profiling Programs" in IBM Parallel Environment for
AIX: Operation and Use, Volume 2, Tools Reference
- -ip
- specifies that the executable is bound with the UDP/IP message passing
support library.
- -us
- specifies that the executable is bound with the RS/6000 SP User Space
message passing library. Executables using this option should be
compiled on an RS/6000 SP node compatible with the node on which execution
will occur.
DESCRIPTION
The mpcc_chkpt shell script invokes the AIX cc
command. The Partition Manager, message passing interface, and
checkpoint support code are automatically linked in. The script creates
an executable with no shared obects. This executable must be run on a
node of the same machine type and having the same level of system software as
the machine on which the executable is built. The executable is not
binary compatible over changes to the system software.
Flags are passed by mpcc_chkpt to the cc command, so most
of the cc options can be used on the mpcc_chkpt shell
script. Options which would generate shared objects should not be
used.
At execution time, the value specified by the MP_EUILIB
environment variable or the -euilib flag must match the -ip |
-us option specified when this command was run.
ENVIRONMENT VARIABLES
- MP_PREFIX
- sets an alternate path to the scripts library. If not set or NULL,
the standard path /usr/lpp/ppe.poe is used. If this
environment variable is set, then all libraries are prefixed by
$MP_PREFIX/ppe.poe.
EXAMPLES
To compile a C program, enter:
mpcc_chkpt program.c -o program
FILES
When you compile a program using mpcc_chkpt, the following
libraries are automatically included:
- /usr/lpp/ppe.poe/lib/libmpi.a (Message Passing Interface,
collective communication routines)
- /usr/lpp/ppe.poe/lib/libvtd.a (VT tracing subsystem)
- /usr/lpp/ppe.poe/lib/libppe.a (PE common routines)
- /usr/lpp/ppe.poe/lib/libc.a (POE version of libc.a)
- /usr/lpp/LoadL/nfs/lib/chkrst.a (LoadLeveler checkpoint/support)
When the -us option is selected, the following libraries are
included:
- /usr/lpp/ppe.poe/lib/us/libmpci.a (PSSP message passing
interface)
- /usr/lpp/ssp/css/lib/hal.a (PSSP User Space adapter interface)
- /usr/lib/swclock.o (PSSP Switch clock interface)
When the -ip option is selected, the following libraries are
included:
- /usr/lpp/ppe.poe/lib/ip/libmpci.a (PSSP message passing
interface)
RELATED INFORMATION
Commands: mpcc(1), mpCC_chkpt(1),
mpxlf_chkpt(1)
NAME
mpcc_r - Invokes a shell script to compile C programs which
use threaded MPI.
SYNOPSIS
mpcc_r [cc_flags]... program.c
The mpcc_r shell script compiles C programs while linking in the
Partition Manager, the threaded implementation of Message Passing Interface
(MPI), and Low-level Applications Programming Interface (LAPI).
FLAGS
Any of the compiler flags normally accepted by the xlc_r or
cc_r command can also be used on mpcc_r. For a
complete listing of these flag options, refer to the manual page for the AIX
cc_r command. Typical options to mpcc_r
include:
- -v
- causes a "verbose" output listing of the shell script.
- -g
- Produces an object file with symbol table references. This object
file is needed for debugging with the pdbx and pedb
debuggers, and is also needed by the Source Code view of the
Visualization Tool (vt).
- -o
- names the executable.
- -d7
- compiles the program with POSIX Threads Draft 7 base MPI and compatibility
libraries. Otherwise, POSIX Threads Draft 10 base libraries are
used.
- -l (lower-case L)
- names additional libraries to be searched. Several libraries are
automatically included, and are listed below in the FILES section.
| Note: | Not all AIX libraries are thread safe. Verify that your intended use
is supported.
|
- -I (upper-case i)
- names directories for additional includes. The directory
/usr/lpp/ppe.poe/include is automatically included.
- -p
- enables profiling with the prof command. For more
information, see the appendix on "Profiling Programs" in IBM
Parallel Environment for AIX: Operation and Use, Volume 2, Tools
Reference
- -pg
- enables profiling with the xprofiler and gprof
commands. For more information, see the "Xprofiler" chapter and
the appendix on "Profiling Programs" in IBM Parallel Environment for
AIX: Operation and Use, Volume 2, Tools Reference
DESCRIPTION
The mpcc_r shell script calls the AIX xlc_r
compiler. In addition, the Partition Manager and data communication
interfaces are automatically linked in. The script creates an
executable that dynamically binds with the communication subsystem
libraries. If you wish to create a statically bound application, use
the instructions in "Creating a Static Executable" in place of this script.
Flags are passed by mpcc_r to the xlc_r command, so any
of the xlc_r options can be used on the mpcc_r shell
script. The communication subsystem library implementation is
dynamically linked when you invoke the executable using the poe
command. The value specified by the MP_EUILIB environment
variable or the -euilib flag will then determine which communication
subsystem library implementation is dynamically linked.
ENVIRONMENT VARIABLES
- MP_PREFIX
- sets an alternate path to the scripts library. If not set or NULL,
the standard path /usr/lpp/ppe.poe is used. If this
environment variable is set, then all libraries are prefixed by
$MP_PREFIX/ppe.poe.
EXAMPLES
To compile a C program, enter:
mpcc_r program.c -o program
FILES
When you compile a program using mpcc_r, the following libraries
are automatically selected:
- /usr/lpp/ppe.poe/lib/libmpi_r.a (Message Passing Interface,
collective communication routines)
- /usr/lpp/ppe.poe/lib/libvtd_r.a (VT tracing subsystem)
- /usr/lpp/ppe.poe/lib/libppe_r.a (PE common routines)
- /usr/lpp/ppe.poe/lib/libc_r.a (POE version of
libc_r.a)
- The following library is selected if it exists as a symbolic link to
/usr/lpp/ssp/css/lib/liblapi_r.a:
/usr/lib/liblapi_r.a
RELATED INFORMATION
Commands: mpCC(1), mpCC_r(1), mpcc(1),
cc(1), pdbx(1), pedb(1), vt(1),
xprofiler(1)
NAME
mpCC - Invokes a shell script to compile C++
programs.
SYNOPSIS
mpCC [xlC_flags]... program.C
The mpCC shell script compiles C++ programs while linking in the
Partition Manager, Message Passing Interface (MPI), and/or Message Passing
Library (MPL).
FLAGS
Any of the compiler flags normally accepted by the xlC command can
also be used on mpCC. For a complete listing of these flag
options, refer to the manual page for the xlC command. Typical
options to mpCC include:
- -v
- causes a "verbose" output listing of the shell script.
- -g
- Produces an object file with symbol table references. This object
file is needed by the Source Code view of the Visualization Tool
(vt).
- -o
- names the executable.
- -l (lower-case L)
- names additional libraries to be searched. Several libraries are
automatically included, and are listed below in the FILES section.
- -I (upper-case i)
- names directories for additional includes. The directory
/usr/lpp/ppe.poe/include is automatically included.
- -p
- enables profiling with the prof command. For more
information, see the appendix on "Profiling Programs" in IBM
Parallel Environment for AIX: Operation and Use, Volume 2, Tools
Reference
- -pg
- enables profiling with the xprofiler and gprof
commands. For more information, see the "Xprofiler" chapter and
the appendix on "Profiling Programs" in IBM Parallel Environment for
AIX: Operation and Use, Volume 2, Tools Reference
DESCRIPTION
The mpCC shell script calls the AIX xlC compiler.
In addition, the Partition Manager and message passing interface are
automatically linked in. The script creates an executable that
dynamically binds with the message passing libraries. If you wish to
create a statically bound application, use the instructions in "Creating a Static Executable" in place of this script.
Flags are passed by mpCC to the xlC command, so any of
the xlC options can be used on the mpCC shell script.
The communication subsystem library implementation is dynamically linked when
you invoke the executable using the poe command. The value
specified by the MP_EUILIB environment variable or the
-euilib flag will then determine which communication subsystem
library implementation is dynamically linked.
ENVIRONMENT VARIABLES
- MP_PREFIX
- sets an alternate path to the scripts library. If not set or NULL,
the standard path /usr/lpp/ppe.poe is used. If this
environment variable is set, then all libraries are prefixed by
$MP_PREFIX/ppe.poe.
EXAMPLES
To compile a C++ program, enter:
mpCC program.C -o program
FILES
When you compile a program using mpCC, the following libraries are
automatically selected:
- /usr/lpp/ppe.poe/lib/libmpi.a (Message passing interface,
collective communication routines)
- /usr/lpp/ppe.poe/lib/libvtd.a (VT tracing subsystem)
- /usr/lpp/ppe.poe/lib/libppe.a (PE common routines)
- /usr/lpp/ppe.poe/lib/libc.a (POE version of libc.a)
RELATED INFORMATION
Commands: mpCC_r(1), mpcc(1), mpcc_r(1),
mpxlf(1), xlC(1), pdbx(1), pedb(1),
vt(1), xprofiler(1)
NAME
mpCC_chkpt - Invokes a shell script to compile
checkpointable C++ programs.
SYNOPSIS
mpCC_chkpt [xlC_flags]... -us | -ip program.C
The mpCC_chkpt shell script compiles C++ programs while linking in
the Partition Manager, Message Passing Interface (MPI), and support code for
parallel Checkpoint/Restart. It builds an executable with no shared
objects.
FLAGS
Most of the compiler flags normally accepted by the xlC command
can also be used on mpCC_chkpt. For a complete listing of
these flag options, refer to the manual page for the AIX xlC
command. Typical options to mpCC_chkpt include:
- -v
- causes a "verbose" output listing of the shell script.
- -g
- Produces an object file with symbol table references. This object
file is needed for debugging with the pdbx and pedb
debuggers, and is also needed by the Source Code view of the
Visualization Tool (vt).
- -o
- names the executable.
- -l (lower-case L)
- names additional libraries to be searched. Several libraries are
automatically included, and are listed below in the FILES section.
- -I (upper-case i)
- names directories for additional includes. The directory
/usr/lpp/ppe.poe/include is automatically included.
- -p
- enables profiling with the prof command. For more
information, see the appendix on "Profiling Programs" in IBM
Parallel Environment for AIX: Operation and Use, Volume 2, Tools
Reference
- -pg
- enables profiling with the xprofiler and gprof
commands. For more information, see the "Xprofiler" chapter and
the appendix on "Profiling Programs" in IBM Parallel Environment for
AIX: Operation and Use, Volume 2, Tools Reference
- -ip
- specifies that the executable is bound with the UDP/IP message passing
support library.
- -us
- specifies that the executable is bound with the RS/6000 SP User Space
message passing library. Executables using this option should be
compiled on an RS/6000 SP node compatible with the node on which execution
will occur.
DESCRIPTION
The mpCC_chkpt shell script invokes the AIX xlC
command. The Partition Manager, message passing interface, and
checkpoint support code are automatically linked in. The script creates
an executable with no shared obects. This executable must be run on a
node of the same machine type and having the same level of system software as
the machine on which the executable is built. The executable is not
binary compatible over changes to the system software.
Flags are passed by mpCC_chkpt to the xlC command, so
most of the xlC options can be used on the mpCC_chkpt shell
script. Options which would generate shared objects should not be
used.
At execution time, the value specified by the MP_EUILIB
environment variable or the -euilib flag must match the -ip |
-us option specified when this command was run.
ENVIRONMENT VARIABLES
- MP_PREFIX
- sets an alternate path to the scripts library. If not set or NULL,
the standard path /usr/lpp/ppe.poe is used. If this
environment variable is set, then all libraries are prefixed by
$MP_PREFIX/ppe.poe.
EXAMPLES
To compile a C++ program, enter:
mpCC_chkpt program.C -o program
FILES
When you compile a program using mpCC_chkpt, the following
libraries are automatically included:
- /usr/lpp/ppe.poe/lib/libmpi.a (Message Passing Interface,
collective communication routines)
- /usr/lpp/ppe.poe/lib/libvtd.a (VT tracing subsystem)
- /usr/lpp/ppe.poe/lib/libppe.a (PE common routines)
- /usr/lpp/ppe.poe/lib/libc.a (POE version of libc.a)
- /usr/lpp/LoadL/full/lib/chkrst.a (LoadLeveler checkpoint/support)
When the -us option is selected, the following libraries are
included:
- /usr/lpp/ppe.poe/lib/us/libmpci.a (PSSP message passing
interface)
- /usr/lpp/ssp/css/lib/hal.a (PSSP User Space adapter interface)
- /usr/lib/swclock.o (PSSP Switch clock interface)
When the -ip option is selected, the following libraries are
included:
- /usr/lpp/ppe.poe/lib/ip/libmpci.a (PSSP message passing
interface)
RELATED INFORMATION
Commands: mpCC(1), mpcc_chkpt(1),
mpxlf_chkpt(1)
NAME
mpCC_r - Invokes a shell script to compile C++ programs
which use threaded MPI.
SYNOPSIS
mpCC_r [xlC_flags]... program.C
The mpCC_r shell script compiles C++ programs while linking in the
Partition Manager, the threaded implementation of Message Passing Interface
(MPI), and Low-level Applications Programming Interface (LAPI).
FLAGS
Any of the compiler flags normally accepted by the xlC_r command
can also be used on mpCC_r. For a complete listing of these
flag options, refer to the manual page for the xlC_r command.
Typical options to mpCC_r include:
- -v
- causes a "verbose" output listing of the shell script.
- -g
- Produces an object file with symbol table references. This object
file is needed by the Source Code view of the Visualization Tool
(vt).
- -o
- names the executable.
- -d7
- compiles the program with POSIX Threads Draft 7 base MPI and compatibility
libraries. Otherwise, POSIX Threads Draft 10 base libraries are
used.
- -l (lower-case L)
- names additional libraries to be searched. Several libraries are
automatically included, and are listed below in the FILES section.
| Note: | Not all AIX libraries are thread safe. Verify that your intended use
is supported.
|
- -I (upper-case i)
- names directories for additional includes. The directory
/usr/lpp/ppe.poe/include is automatically included.
- -p
- enables profiling with the prof command. For more
information, see the appendix on "Profiling Programs" in IBM
Parallel Environment for AIX: Operation and Use, Volume 2, Tools
Reference
- -pg
- enables profiling with the xprofiler and gprof
commands. For more information, see the "Xprofiler" chapter and
the appendix on "Profiling Programs" in IBM Parallel Environment for
AIX: Operation and Use, Volume 2, Tools Reference
DESCRIPTION
The mpCC_r shell script calls the AIX xlC_r
compiler. In addition, the Partition Manager and data communication
interfaces are automatically linked in. The script creates an
executable that dynamically binds with the communication subsystem
libraries. If you wish to create a statically bound application, use
the instructions in "Creating a Static Executable" in place of this script.
Flags are passed by mpCC_r to the xlC_r command, so any
of the xlC_r options can be used on the mpCC_r shell
script. The communication subsystem library implementation is
dynamically linked when you invoke the executable using the poe
command. The value specified by the MP_EUILIB environment
variable or the -euilib flag will then determine which communication
subsystem library implementation is dynamically linked.
ENVIRONMENT VARIABLES
- MP_PREFIX
- sets an alternate path to the scripts library. If not set or NULL,
the standard path /usr/lpp/ppe.poe is used. If this
environment variable is set, then all libraries are prefixed by
$MP_PREFIX/ppe.poe.
EXAMPLES
To compile a C++ program, enter:
mpCC_r program.C -o program
FILES
When you compile a program using mpCC_r, the following libraries
are automatically selected:
- /usr/lpp/ppe.poe/lib/libmpi_r.a (Message passing interface,
collective communication routines)
- /usr/lpp/ppe.poe/lib/libvtd_r.a (VT tracing subsystem)
- /usr/lpp/ppe.poe/lib/libppe_r.a (PE common routines)
- /usr/lpp/ppe.poe/lib/libc_r.a (POE version of
libc_r.a)
- The following library is selected if it exists as a symbolic link to
/usr/lpp/ssp/css/lib/liblapi_r.a:
/usr/lib/liblapi_r.a
RELATED INFORMATION
Commands: mpcc_r(1), mpcc(1), mpCC(1),
xlC(1), pdbx(1), pedb(1), vt(1),
xprofiler(1)
NAME
mprcp - copies a file from the home node to a list of remote
hosts.
SYNOPSIS
mprcp host_list_file file_id
The mprcp shell script uses the rcp command to copy the
file specified by file_id to all the remote hosts listed in the
specified host_list_file. This file_id must be
specified by an absolute path name.
FLAGS
None.
DESCRIPTION
The mprcp shell script is typically used to distribute
executables, data files, and (in order to use the parallel debuggers) source
code files from the home node to all the remote nodes of the partition prior
to invoking poe. This only needs to be done if the needed
files are not in a shared file system, or are not part of a file collection
which is distributed automatically.
For each remote host listed in the specified host_list_file, the
size and date of file_id on the remote system is determined.
If the file does not exist on the remote system, or if either the size or date
differs from the corresponding statistic for the local system, the
rcp command is used to copy the file to the remote system.
The copied file retains the local system size and date and overlays any
existing file of the same name. The remote copy uses the same userid as
the local system. If the remote host cannot be contacted, or if the
rcp command fails, an error message is printed and the script
exits.
EXAMPLES
To send a copy of the executable program to all the processor
nodes listed in host.list in your current directory,
enter:
mprcp host.list $PWD/program
RELATED INFORMATION
Commands: rcp(1), mcp(1)
NAME
mpxlf - Invokes a shell script to compile Fortran
programs.
SYNOPSIS
mpxlf [xlf_flags]... program.f
The mpxlf shell script compiles Fortran programs while linking in
the Partition Manager, Message Passing Interface (MPI), and/or Message Passing
Library (MPL).
FLAGS
Any of the compiler flags normally accepted by the xlf command can
also be used on mpxlf. For a complete listing of these flag
options, refer to the manual page for the xlf command. Typical
options to mpxlf include:
- -v
- causes a "verbose" output listing of the shell script.
- -g
- Produces an object file with symbol table references. This object
file is needed for debugging with the pdbx and pedb
debuggers, and is also needed by the Source Code view of the Visualization
Tool.
- -o
- names the executable.
- -l (lower-case L)
- names additional libraries to be searched. Several libraries are
automatically included, and are listed below in the FILES section.
- -I (upper-case i)
- names directories for additional includes. The directory
/usr/lpp/ppe.poe/include is automatically included.
- -p
- enables profiling with the prof command. For more
information, see the appendix on "Profiling Programs" in IBM
Parallel Environment for AIX: Operation and Use, Volume 2, Tools
Reference
- -pg
- enables profiling with the xprofiler and gprof
commands. For more information, see the "Xprofiler" chapter and
the appendix on "Profiling Programs" in IBM Parallel Environment for
AIX: Operation and Use, Volume 2, Tools Reference
DESCRIPTION
The mpxlf shell script calls the xlf compiler. In
addition, the Partition Manager and message passing interface are
automatically linked in. The script creates an executable that
dynamically binds with the message passing libraries. If you wish to
create a statically bound application, use the instructions in "Creating a Static Executable" in place of this script.
Flags are passed by mpxlf to the xlf command, so any of
the xlf options can be used on the mpxlf shell
script. The communication subsystem library implementation is
dynamically linked when you invoke the executable using the poe
command. The value specified by the MP_EUILIB environment
variable or the -euilib flag will then determine which communication
subsystem library implementation is dynamically linked.
mpxlf, mpxlf_r, mpxlf90, mpxlf90_r,
mpxlf_chkpt, mpxlf90_chkpt have been updated to support High
Performance Fortran (HPF) Version 1, Release 3. In doing so, if other
Fortran compilers are installed on your system in addition to HPF 1.3,
you may need to use a new environment variable with the compiler script in
order to use HPF 1.3.
xlf and HPF use different compiler paths and stanzas.
Therefore, the scripts will now check for the compiler level installed, and
will do the following:
- If only xlf is installed on the system, it will be used.
- If only HPF is installed, only HPF 1.3 or greater is supported and
it will be used.
- If both xlf and HPF 1.3 are installed, xlf is
used as the default, unless the customer overrides it by specifying the
MP_HPF environment variable.
As such, the POE Fortran compile scripts check for a new environment
variable, MP_HPF, to determine if the HPF should be used.
Customers with both HPF and xlf installed should set
MP_HPF=YES when they desire to use the HPF 1.3
compiler.
Customers without both HPF and xlf installed do not need to set
the MP_HPF variable.
ENVIRONMENT VARIABLES
- MP_PREFIX
- sets an alternate path to the scripts library. If not set or NULL,
the standard path /usr/lpp/ppe.poe is used. If this
environment variable is set, then all libraries are prefixed by
$MP_PREFIX/ppe.poe.
- MP_HPF
- If High Performance Fortran (HPF) 1.3 is installed along with other
Fortran compilers, set this to YES if the HPF 1.3 compiler is
to be used. Otherwise the default xlf compiler will be
used.
EXAMPLES
To compile a Fortran program, enter:
mpxlf program.f -o program
FILES
When you compile a program using mpxlf, the following libraries
are automatically selected:
- /usr/lpp/ppe.poe/lib/libmpi.a (Message passing interface,
collective communication routines)
- /usr/lpp/ppe.poe/lib/libvtd.a (VT tracing subsystem)
- /usr/lpp/ppe.poe/lib/libppe.a (PE common routines)
- /usr/lpp/ppe.poe/lib/libc.a (POE version of libc.a)
RELATED INFORMATION
Commands: mpcc(1), xlf(1), pdbx(1),
pedb(1), vt(1), xprofiler(1)
NAME
mpxlf_chkpt - Invokes a shell script to compile
checkpointable FORTRAN programs.
SYNOPSIS
mpxlf_chkpt [xlf_flags]... -us | -ip program.f
The mpxlf_chkpt shell script compiles FORTRAN programs while
linking in the Partition Manager, Message Passing Interface (MPI), and support
code for parallel Checkpoint/Restart. It builds an executable with no
shared objects.
FLAGS
Most of the compiler flags normally accepted by the xlf command
can also be used on mpxlf_chkpt. For a complete listing of
these flag options, refer to the manual page for the AIX xlf
command. Typical options to mpxlf_chkpt include:
- -v
- causes a "verbose" output listing of the shell script.
- -g
- Produces an object file with symbol table references. This object
file is needed for debugging with the pdbx and pedb
debuggers, and is also needed by the Source Code view of the
Visualization Tool (vt).
- -o
- names the executable.
- -l (lower-case L)
- names additional libraries to be searched. Several libraries are
automatically included, and are listed below in the FILES section.
- -I (upper-case i)
- names directories for additional includes. The directory
/usr/lpp/ppe.poe/include is automatically included.
- -p
- enables profiling with the prof command. For more
information, see the appendix on "Profiling Programs" in IBM
Parallel Environment for AIX: Operation and Use, Volume 2, Tools
Reference
- -pg
- enables profiling with the xprofiler and gprof
commands. For more information, see the "Xprofiler" chapter and
the appendix on "Profiling Programs" in IBM Parallel Environment for
AIX: Operation and Use, Volume 2, Tools Reference
- -ip
- specifies that the executable is bound with the UDP/IP message passing
support library.
- -us
- specifies that the executable is bound with the RS/6000 SP User Space
message passing library. Executables using this option should be
compiled on an RS/6000 SP node compatible with the node on which execution
will occur.
DESCRIPTION
The mpxlf_chkpt shell script invokes the AIX xlf
command. The Partition Manager, message passing interface, and
checkpoint support code are automatically linked in. The script creates
an executable with no shared obects. This executable must be run on a
node of the same machine type and having the same level of system software as
the machine on which the executable is built. The executable is not
binary compatible over changes to the system software.
Flags are passed by mpxlf_chkpt to the xlf command, so
most of the xlf options can be used on the mpxlf_chkpt shell
script. Options which would generate shared objects should not be
used.
At execution time, the value specified by the MP_EUILIB
environment variable or the -euilib flag must match the -ip |
-us option specified when this command was run.
mpxlf, mpxlf_r, mpxlf90, mpxlf90_r,
mpxlf_chkpt, mpxlf90_chkpt have been updated to support High
Performance Fortran (HPF) Version 1, Release 3. In doing so, if other
Fortran compilers are installed on your system in addition to HPF 1.3,
you may need to use a new environment variable with the compiler script in
order to use HPF 1.3.
xlf and HPF use different compiler paths and stanzas.
Therefore, the scripts will now check for the compiler level installed, and
will do the following:
- If only xlf is installed on the system, it will be used.
- If only HPF is installed, only HPF 1.3 or greater is supported and
it will be used.
- If both xlf and HPF 1.3 are installed, xlf is
used as the default, unless the customer overrides it by specifying the
MP_HPF environment variable.
As such, the POE Fortran compile scripts check for a new environment
variable, MP_HPF, to determine if the HPF should be used.
Customers with both HPF and xlf installed should set
MP_HPF=YES when they desire to use the HPF 1.3
compiler.
Customers without both HPF and xlf installed do not need to set
the MP_HPF variable.
ENVIRONMENT VARIABLES
- MP_PREFIX
- sets an alternate path to the scripts library. If not set or NULL,
the standard path /usr/lpp/ppe.poe is used. If this
environment variable is set, then all libraries are prefixed by
$MP_PREFIX/ppe.poe.
- MP_HPF
- If High Performance Fortran (HPF) 1.3 is installed along with other
Fortran compilers, set this to YES if the HPF 1.3 compiler is
to be used. Otherwise the default xlf compiler will be
used.
EXAMPLES
To compile a FORTRAN program, enter:
mpxlf_chkpt program.f -o program
FILES
When you compile a program using mpxlf_chkpt, the following
libraries are automatically included:
- /usr/lpp/ppe.poe/lib/libmpi.a (Message Passing Interface,
collective communication routines)
- /usr/lpp/ppe.poe/lib/libvtd.a (VT tracing subsystem)
- /usr/lpp/ppe.poe/lib/libppe.a (PE common routines)
- /usr/lpp/ppe.poe/lib/libc.a (POE version of libc.a)
- /usr/lpp/LoadL/full/lib/chkrst.a (LoadLeveler checkpoint/support)
When the -us option is selected, the following libraries are
included:
- /usr/lpp/ppe.poe/lib/us/libmpci.a (PSSP message passing
interface)
- /usr/lpp/ssp/css/lib/hal.a (PSSP User Space adapter interface)
- /usr/lib/swclock.o (PSSP Switch clock interface)
When the -ip option is selected, the following libraries are
included:
- /usr/lpp/ppe.poe/lib/ip/libmpci.a (PSSP message passing
interface)
RELATED INFORMATION
Commands: mpxlf(1), mpcc_chkpt(1),
mpxlf90_chkpt(1)
NAME
mpxlf_r - Invokes a shell script to compile Fortran programs
and link them into a threaded environment.
SYNOPSIS
mpxlf_r [xlf_flags]... program.f
The mpxlf_r shell script compiles Fortran programs while linking
in the Partition Manager, the threaded implementation of Message Passing
Interface (MPI), and Low-level Applications Programming Interface
(LAPI).
| Note: | Only one thread can run a Fortran program.
|
FLAGS
Any of the compiler flags normally accepted by the xlf command can
also be used on mpxlf_r. For a complete listing of these flag
options, refer to the manual page for the xlf command. Typical
options to mpxlf_r include:
- -v
- causes a "verbose" output listing of the shell script.
- -g
- Produces an object file with symbol table references. This object
file is needed for debugging with the pdbx and pedb
debuggers, and is also needed by the Source Code view of the Visualization
Tool.
- -o
- names the executable.
- -d7
- compiles the program with POSIX Threads Draft 7 base MPI and compatibility
libraries. Otherwise, POSIX Threads Draft 10 base libraries are
used. This flag can only be used with Fortran Version
5.1.1 or later.
- -l (lower-case L)
- names additional libraries to be searched. Several libraries are
automatically included, and are listed below in the FILES section.
| Note: | Not all AIX libraries are thread safe. Verify that your intended use
is supported.
|
- -I (upper-case i)
- names directories for additional includes. The directory
/usr/lpp/ppe.poe/include is automatically included.
- -p
- enables profiling with the prof command. For more
information, see the appendix on "Profiling Programs" in IBM
Parallel Environment for AIX: Operation and Use, Volume 2, Tools
Reference
- -pg
- enables profiling with the xprofiler and gprof
commands. For more information, see the "Xprofiler" chapter and
the appendix on "Profiling Programs" in IBM Parallel Environment for
AIX: Operation and Use, Volume 2, Tools Reference
DESCRIPTION
The mpxlf_r shell script calls the xlf compiler.
In addition, the Partition Manager and data communication interfaces are
automatically linked in. The script creates an executable that
dynamically binds with the communication subsystem libraries. If you
wish to create a statically bound application, use the instructions in "Creating a Static Executable" in place of this script.
Flags are passed by mpxlf_r to the xlf command, so any of
the xlf options can be used on the mpxlf_r shell
script. The communication subsystem library implementation is
dynamically linked when you invoke the executable using the poe
command. The value specified by the MP_EUILIB environment
variable or the -euilib flag will then determine which communication
subsystem library implementation is dynamically linked.
mpxlf, mpxlf_r, mpxlf90, mpxlf90_r,
mpxlf_chkpt, mpxlf90_chkpt have been updated to support High
Performance Fortran (HPF) Version 1, Release 3. In doing so, if other
Fortran compilers are installed on your system in addition to HPF 1.3,
you may need to use a new environment variable with the compiler script in
order to use HPF 1.3.
xlf and HPF use different compiler paths and stanzas.
Therefore, the scripts will now check for the compiler level installed, and
will do the following:
- If only xlf is installed on the system, it will be used.
- If only HPF is installed, only HPF 1.3 or greater is supported and
it will be used.
- If both xlf and HPF 1.3 are installed, xlf is
used as the default, unless the customer overrides it by specifying the
MP_HPF environment variable.
As such, the POE Fortran compile scripts check for a new environment
variable, MP_HPF, to determine if the HPF should be used.
Customers with both HPF and xlf installed should set
MP_HPF=YES when they desire to use the HPF 1.3
compiler.
Customers without both HPF and xlf installed do not need to set
the MP_HPF variable.
ENVIRONMENT VARIABLES
- MP_PREFIX
- sets an alternate path to the scripts library. If not set or NULL,
the standard path /usr/lpp/ppe.poe is used. If this
environment variable is set, then all libraries are prefixed by
$MP_PREFIX/ppe.poe.
- MP_HPF
- If High Performance Fortran (HPF) 1.3 is installed along with other
Fortran compilers, set this to YES if the HPF 1.3 compiler is
to be used. Otherwise the default xlf compiler will be
used.
EXAMPLES
To compile a Fortran program, enter:
mpxlf_r program.f -o program
FILES
When you compile a program using mpxlf_r, the following libraries
are automatically selected:
- /usr/lpp/ppe.poe/lib/libmpi_r.a (Message passing interface,
collective communication routines)
- /usr/lpp/ppe.poe/lib/libvtd_r.a (VT tracing subsystem)
- /usr/lpp/ppe.poe/lib/libppe_r.a (PE common routines)
- /usr/lpp/ppe.poe/lib/libc_r.a (POE version of
libc_r.a)
- The following library is selected if it exists as a symbolic link to
/usr/lpp/ssp/css/lib/liblapi_r.a:
/usr/lib/liblapi_r.a
| Note: | Fortran Version 4.1.0.1, specifically the library
libxlf90_t.a, must be available for both linking and execution.
|
RELATED INFORMATION
Commands: mpcc_r(1), xlf(1), pdbx(1),
pedb(1), vt(1), xprofiler(1)
NAME
mpxlf90 - Invokes a shell script to compile Fortran 90
programs.
SYNOPSIS
mpxlf90 [xlf_flags]... program.f
The mpxlf90 shell script compiles Fortran 90 programs while
linking in the Partition Manager, Message Passing Interface (MPI), and Message
Passing Library (MPL).
FLAGS
Any of the compiler flags normally accepted by the xlf command can
also be used on mpxlf90. For a complete listing of these flag
options, refer to the manual page for the xlf command. Typical
options to mpxlf90 include:
- -v
- causes a "verbose" output listing of the shell script.
- -g
- Produces an object file with symbol table references. This object
file is needed by the Source Code view of the Visualization Tool.
- -o
- names the executable.
- -l (lower-case L)
- names additional libraries to be searched. Several libraries are
automatically included, and are listed below in the FILES section.
- -I (upper-case i)
- names directories for additional includes. The directory
/usr/lpp/ppe.poe/include is automatically included.
- -p
- enables profiling with the prof command. For more
information, see the appendix on "Profiling Programs" in IBM
Parallel Environment for AIX: Operation and Use, Volume 2, Tools
Reference
- -pg
- enables profiling with the xprofiler and gprof
commands. For more information, see the "Xprofiler" chapter and
the appendix on "Profiling Programs" in IBM Parallel Environment for
AIX: Operation and Use, Volume 2, Tools Reference
DESCRIPTION
The mpxlf90 shell script calls the xlf compiler.
In addition, the Partition Manager and message passing interface are
automatically linked in. The script creates an executable that
dynamically binds with the message passing libraries. If you wish to
create a statically bound application, use the instructions in "Creating a Static Executable" in place of this script.
Flags are passed by mpxlf90 to the xlf command, so any of
the xlf options can be used on the mpxlf90 shell
script. The communication subsystem library implementation is
dynamically linked when you invoke the executable using the poe
command. The value specified by the MP_EUILIB environment
variable or the -euilib flag will then determine which communication
subsystem library implementation is dynamically linked.
mpxlf, mpxlf_r, mpxlf90, mpxlf90_r,
mpxlf_chkpt, mpxlf90_chkpt have been updated to support High
Performance Fortran (HPF) Version 1, Release 3. In doing so, if other
Fortran compilers are installed on your system in addition to HPF 1.3,
you may need to use a new environment variable with the compiler script in
order to use HPF 1.3.
xlf and HPF use different compiler paths and stanzas.
Therefore, the scripts will now check for the compiler level installed, and
will do the following:
- If only xlf is installed on the system, it will be used.
- If only HPF is installed, only HPF 1.3 or greater is supported and
it will be used.
- If both xlf and HPF 1.3 are installed, xlf is
used as the default, unless the customer overrides it by specifying the
MP_HPF environment variable.
As such, the POE Fortran compile scripts check for a new environment
variable, MP_HPF, to determine if the HPF should be used.
Customers with both HPF and xlf installed should set
MP_HPF=YES when they desire to use the HPF 1.3
compiler.
Customers without both HPF and xlf installed do not need to set
the MP_HPF variable.
ENVIRONMENT VARIABLES
- MP_PREFIX
- sets an alternate path to the scripts library. If not set or NULL,
the standard path /usr/lpp/ppe.poe is used. If this
environment variable is set, then all libraries are prefixed by
$MP_PREFIX/ppe.poe.
- MP_HPF
- If High Performance Fortran (HPF) 1.3 is installed along with other
Fortran compilers, set this to YES if the HPF 1.3 compiler is
to be used. Otherwise the default xlf compiler will be
used.
EXAMPLES
To compile a Fortran 90 program, enter:
mpxlf90 program.f -o program
FILES
When you compile a program using mpxlf90, the following libraries
are automatically selected:
- /usr/lpp/ppe.poe/lib/libmpi.a (Message passing interface,
collective communication routines)
- /usr/lpp/ppe.poe/lib/libvtd.a (VT tracing subsystem)
- /usr/lpp/ppe.poe/lib/libppe.a (PE common routines)
- /usr/lpp/ppe.poe/lib/libc.a (POE version of libc.a)
RELATED INFORMATION
Commands: mpcc(1), mpCC(1), xlf(1),
mpxlf(1), vt(1), xprofiler(1)
NAME
mpxlf90_chkpt - Invokes a shell script to compile
checkpointable Fortran 90 programs.
SYNOPSIS
mpxlf90_chkpt [xlf90_flags]... -us | -ip program.f
The mpxlf90_chkpt shell script compiles Fortran 90 programs while
linking in the Partition Manager, Message Passing Interface (MPI), and support
code for parallel Checkpoint/Restart. It builds an executable with no
shared objects.
FLAGS
Most of the compiler flags normally accepted by the xlf90 command
can also be used on mpxlf90_chkpt. For a complete listing of
these flag options, refer to the manual page for the AIX xlf90
command. Typical options to mpxlf90_chkpt include:
- -v
- causes a "verbose" output listing of the shell script.
- -g
- Produces an object file with symbol table references. This object
file is needed for debugging with the pdbx and pedb
debuggers, and is also needed by the Source Code view of the
Visualization Tool (vt).
- -o
- names the executable.
- -l (lower-case L)
- names additional libraries to be searched. Several libraries are
automatically included, and are listed below in the FILES section.
- -I (upper-case i)
- names directories for additional includes. The directory
/usr/lpp/ppe.poe/include is automatically included.
- -p
- enables profiling with the prof command. For more
information, see the appendix on "Profiling Programs" in IBM
Parallel Environment for AIX: Operation and Use, Volume 2, Tools
Reference
- -pg
- enables profiling with the xprofiler and gprof
commands. For more information, see the "Xprofiler" chapter and
the appendix on "Profiling Programs" in IBM Parallel Environment for
AIX: Operation and Use, Volume 2, Tools Reference
- -ip
- specifies that the executable is bound with the UDP/IP message passing
support library.
- -us
- specifies that the executable is bound with the RS/6000 SP User Space
message passing library. Executables using this option should be
compiled on an RS/6000 SP node compatible with the node on which execution
will occur.
DESCRIPTION
The mpxlf90_chkpt shell script invokes the AIX xlf90
command. The Partition Manager, message passing interface, and
checkpoint support code are automatically linked in. The script creates
an executable with no shared obects. This executable must be run on a
node of the same machine type and having the same level of system software as
the machine on which the executable is built. The executable is not
binary compatible over changes to the system software.
Flags are passed by mpxlf90_chkpt to the xlf90 command,
so most of the xlf90 options can be used on the
mpxlf90_chkpt shell script. Options which would generate
shared objects should not be used.
At execution time, the value specified by the MP_EUILIB
environment variable or the -euilib flag must match the -ip |
-us option specified when this command was run.
mpxlf, mpxlf_r, mpxlf90, mpxlf90_r,
mpxlf_chkpt, mpxlf90_chkpt have been updated to support High
Performance Fortran (HPF) Version 1, Release 3. In doing so, if other
Fortran compilers are installed on your system in addition to HPF 1.3,
you may need to use a new environment variable with the compiler script in
order to use HPF 1.3.
xlf and HPF use different compiler paths and stanzas.
Therefore, the scripts will now check for the compiler level installed, and
will do the following:
- If only xlf is installed on the system, it will be used.
- If only HPF is installed, only HPF 1.3 or greater is supported and
it will be used.
- If both xlf and HPF 1.3 are installed, xlf is
used as the default, unless the customer overrides it by specifying the
MP_HPF environment variable.
As such, the POE Fortran compile scripts check for a new environment
variable, MP_HPF, to determine if the HPF should be used.
Customers with both HPF and xlf installed should set
MP_HPF=YES when they desire to use the HPF 1.3
compiler.
Customers without both HPF and xlf installed do not need to set
the MP_HPF variable.
ENVIRONMENT VARIABLES
- MP_PREFIX
- sets an alternate path to the scripts library. If not set or NULL,
the standard path /usr/lpp/ppe.poe is used. If this
environment variable is set, then all libraries are prefixed by
$MP_PREFIX/ppe.poe.
- MP_HPF
- If High Performance Fortran (HPF) 1.3 is installed along with other
Fortran compilers, set this to YES if the HPF 1.3 compiler is
to be used. Otherwise the default xlf compiler will be
used.
EXAMPLES
To compile a FORTRAN program, enter:
mpxlf90_chkpt program.f -o program
FILES
When you compile a program using mpxlf90_chkpt, the following
libraries are automatically included:
- /usr/lpp/ppe.poe/lib/libmpi.a (Message Passing Interface,
collective communication routines)
- /usr/lpp/ppe.poe/lib/libvtd.a (VT tracing subsystem)
- /usr/lpp/ppe.poe/lib/libppe.a (PE common routines)
- /usr/lpp/ppe.poe/lib/libc.a (POE version of libc.a)
- /usr/lpp/LoadL/full/lib/chkrst.a (LoadLeveler checkpoint/support)
When the -us option is selected, the following libraries are
included:
- /usr/lpp/ppe.poe/lib/us/libmpci.a (PSSP message passing
interface)
- /usr/lpp/ssp/css/lib/hal.a (PSSP User Space adapter interface)
- /usr/lib/swclock.o (PSSP Switch clock interface)
When the -ip option is selected, the following libraries are
included:
- /usr/lpp/ppe.poe/lib/ip/libmpci.a (PSSP message passing
interface)
RELATED INFORMATION
Commands: mpxlf90(1), mpcc_chkpt(1),
mpxlf_chkpt(1)
NAME
mpxlf90_r - Invokes a shell script to compile Fortran 90
programs and link them into a threaded environment.
SYNOPSIS
mpxlf90_r [xlf_flags]... program.f
The mpxlf90_r shell script compiles Fortran 90 programs while
linking in the Partition Manager, the threaded implementation of Message
Passing Interface (MPI), and Low-level Applications Programming Interface
(LAPI).
| Note: | Only one thread can run a Fortran program.
|
FLAGS
Any of the compiler flags normally accepted by the xlf command can
also be used on mpxlf90_r. For a complete listing of these
flag options, refer to the manual page for the xlf command.
Typical options to mpxlf90_r include:
- -v
- causes a "verbose" output listing of the shell script.
- -g
- Produces an object file with symbol table references. This object
file is needed by the Source Code view of the Visualization Tool.
- -o
- names the executable.
- -d7
- compiles the program with POSIX Threads Draft 7 base MPI and compatibility
libraries. Otherwise, POSIX Threads Draft 10 base libraries are
used. This flag can only be used with Fortran Version
5.1.1 or later.
- -l (lower-case L)
- names additional libraries to be searched. Several libraries are
automatically included, and are listed below in the FILES section.
| Note: | Not all AIX libraries are thread safe. Verify that your intended use
is supported.
|
- -I (upper-case i)
- names directories for additional includes. The directory
/usr/lpp/ppe.poe/include is automatically included.
- -p
- enables profiling with the prof command. For more
information, see the appendix on "Profiling Programs" in IBM
Parallel Environment for AIX: Operation and Use, Volume 2, Tools
Reference
- -pg
- enables profiling with the xprofiler and gprof
commands. For more information, see the "Xprofiler" chapter and
the appendix on "Profiling Programs" in IBM Parallel Environment for
AIX: Operation and Use, Volume 2, Tools Reference
DESCRIPTION
The mpxlf90_r shell script calls the xlf compiler.
In addition, the Partition Manager and data communication interfaces are
automatically linked in. The script creates an executable that
dynamically binds with the communication subsystem libraries. If you
wish to create a statically bound application, use the instructions in "Creating a Static Executable" in place of this script.
Flags are passed by mpxlf90_r to the xlf command, so any
of the xlf options can be used on the mpxlf90_r shell
script. The communication subsystem library implementation is
dynamically linked when you invoke the executable using the poe
command. The value specified by the MP_EUILIB environment
variable or the -euilib flag will then determine which communication
subsystem library implementation is dynamically linked.
mpxlf, mpxlf_r, mpxlf90, mpxlf90_r,
mpxlf_chkpt, mpxlf90_chkpt have been updated to support High
Performance Fortran (HPF) Version 1, Release 3. In doing so, if other
Fortran compilers are installed on your system in addition to HPF 1.3,
you may need to use a new environment variable with the compiler script in
order to use HPF 1.3.
xlf and HPF use different compiler paths and stanzas.
Therefore, the scripts will now check for the compiler level installed, and
will do the following:
- If only xlf is installed on the system, it will be used.
- If only HPF is installed, only HPF 1.3 or greater is supported and
it will be used.
- If both xlf and HPF 1.3 are installed, xlf is
used as the default, unless the customer overrides it by specifying the
MP_HPF environment variable.
As such, the POE Fortran compile scripts check for a new environment
variable, MP_HPF, to determine if the HPF should be used.
Customers with both HPF and xlf installed should set
MP_HPF=YES when they desire to use the HPF 1.3
compiler.
Customers without both HPF and xlf installed do not need to set
the MP_HPF variable.
ENVIRONMENT VARIABLES
- MP_PREFIX
- sets an alternate path to the scripts library. If not set or NULL,
the standard path /usr/lpp/ppe.poe is used. If this
environment variable is set, then all libraries are prefixed by
$MP_PREFIX/ppe.poe.
- MP_HPF
- If High Performance Fortran (HPF) 1.3 is installed along with other
Fortran compilers, set this to YES if the HPF 1.3 compiler is
to be used. Otherwise the default xlf compiler will be
used.
EXAMPLES
To compile a Fortran 90 program, enter:
mpxlf90_r program.f -o program
FILES
When you compile a program using mpxlf90_r, the following
libraries are automatically selected:
- /usr/lpp/ppe.poe/lib/libmpi_r.a (Message passing interface,
collective communication routines)
- /usr/lpp/ppe.poe/lib/libvtd_r.a (VT tracing subsystem)
- /usr/lpp/ppe.poe/lib/libppe_r.a (PE common routines)
- /usr/lpp/ppe.poe/lib/libc_r.a (POE version of
libc_r.a)
- The following library is selected if it exists as a symbolic link to
/usr/lpp/ssp/css/lib/liblapi_r.a:
/usr/lib/liblapi_r.a
| Note: | Fortran Version 4.1.0.1, specifically the library
libxlf90_t.a, must be available for both linking and execution.
|
RELATED INFORMATION
Commands: mpcc_r(1), xlf(1), mpxlf_r(1),
pdbx(1), pedb(1), vt(1), xprofiler(1)
NAME
pmarray - Starts the Program Marker Array, which is an
X-Windows tool for monitoring a parallel executable's run.
SYNOPSIS
pmarray
The pmarray command starts the Program Marker Array X-Windows tool
prior to invoking poe. This tool is used for run-time
monitoring.
FLAGS
None.
DESCRIPTION
The pmarray command starts the Program Marker Array. This
X-Windows run-time monitoring tool consists of a number of small squares, or
lights. Each task in a parallel program has its own row of
lights. Using calls to the Parallel Utility Functions enables a
parallel program to control the appearance of the Program Marker Array
Window. Calls to MP_MARKER (for Fortran programs) or mpc_marker (for C
programs) enables a program to color lights on, and/or send output strings to
the Window. Calls to MP_NLIGHTS (for Fortran programs) and mpc_nlights
(for C programs) enables a program to determine the number of lights displayed
per task row.
ENVIRONMENT VARIABLES
This command responds to the following environment variables:
- MP_PMLIGHTS
- Indicates the number of lights displayed per row on the Array. (If,
when you invoke poe, you override MP_PMLIGHTS using the
-pmlights flag, the
Array will redisplay with the new number of lights per row.)
- MP_PROCS
- Indicates the number of program tasks. The pmarray command
sets the number of task rows displayed in the Array equal to the value of
MP_PROCS. (If, when you invoke poe, you override
MP_PROCS using the -procs flag, the Array will redisplay
with the new number of rows.)
- MP_USRPORT
- Indicates the port id used by the Partition Manager to connect to the
Array. By default, the Partition Manager connects to the Array using a
socket assigned to port 9999. If you get an error message indicating
that the port is in use, specify a different port. Standard TCP/IP
practice suggests using ports greater than 5000 and less than 10000.
EXAMPLES
To start the Program Marker Array program as a background process, and open
its window, enter:
pmarray &
FILES
/usr/lib/X11/app-defaults/PMarray (Xdefaults file)
/usr/lib/ppe.poe/samples/PMarray.ad (X-Windows resource
information)
RELATED INFORMATION
Commands: poe(1)
Subroutines: mpc_marker(3), MP_MARKER(3),
mpc_nlights(3), MP_NLIGHTS(3)
"Monitoring Program Execution Using the Program Marker Array" in
IBM Parallel Environment for AIX: Operation and Use
NAME
poe - Invokes the Parallel Operating Environment (POE) for
loading and executing programs on remote processor nodes.
SYNOPSIS
poe [-h] [program] [program_options]...
[-adapter_use adapter_specifier]
[-cpu_use cpu_specifier]
[-euidevice device_specifier]
[-euilib {ip | us}]
[-euilibpath path_specifier]
[{-hostfile | -hfile} host_file_name]
[-procs partition_size]
[-pulse interval]
[-resd {yes | no}]
[-retry retry_interval]
[-retrycount retry_count]
[-rmpool pool_ID]
[-savehostfile output_file_name]
[-spname name]
[-cmdfile commands_file]
[-newjob {yes | no}]
[-pgmmodel {spmd | mpmd}]
[-labelio {yes | no}]
[-stdinmode {all | none | task_ID}]
[-stdoutmode {unordered | ordered | task_ID}]
[{-samplefreq | -sfreq} samp_int]
[{-tbuffsize | -tbsize} buffer_size]
[{-tbuffwrap | -tbwrap} {yes | no}]
[-tmpdir temp_trace_directory]
[{-tracedir | -tdir} final_trace_directory]
[{-tracefile | -tfile} trace_file_name]
[{-tracelevel | -tlevel} level_int]
[{-ttempsize | -ttsize} temp_file_size]
[{-infolevel | -ilevel} message_level]
[-pmdlog {yes | no}]
[-buffer_mem] memory_size
[-css_interrupt {yes | no}]
[-eager_limit size_limit]
[-intrdelay delay_parameter]
[-max_typedepth maximum_depth]
[-use_flow_control {yes | no}]
[-thread_stacksize]
[-single_thread {no | yes}]
[-wait_mode {poll | yield | sleep}]
[-polling_interval interval]
[-euidevelop {yes | no | deb | min | nor}]
[-pmlights number_of_lights]
[-usrport port_ID] [fence_string additional_options]
[-coredir]
The poe command invokes the Parallel Operating Environment for
loading and executing programs on remote processor nodes. The operation
of POE is influenced by a number of POE environment variables. The flag
options on this command are each used to temporarily override one of these
environment variables. User program_options can be freely
interspersed with the flag options. If no program is
specified, POE will either prompt you for programs to load, or, if the
MP_CMDFILE environment variable is set, will load the partition
using the specified commands file.
FLAGS
The -h flag, when used, must appear immediately after
poe, and causes the poe man page, if it exists, to be
printed to the screen.
The remaining flags you can specify on this command are used to temporarily
override POE environment variables. For more information on valid
values, and on what a particular flag sets, refer to the description of its
associated environment variable in the ENVIRONMENT VARIABLES section.
The following flags are grouped by function.
The following Partition Manager control flags override the associated
environment variables.
- -adapter_use
- MP_ADAPTER_USE
- -cpu_use
- MP_CPU_USE
- -euidevice
- MP_EUIDEVICE
- -euilib
- MP_EUILIB
- -euilibpath
- MP_EUILIBPATH
- -hostfile or -hfile
- MP_HOSTFILE
- -procs
- MP_PROCS
- -pulse
- MP_PULSE
- -resd
- MP_RESD
- -retry
- MP_RETRY
- -retrycount
- MP_RETRYCOUNT
- -msg_api
- MP_MSG_API
- -rmpool
- MP_RMPOOL
- -nodes
- MP_NODES
- -tasks_per_node
- MP_TASKS_PER_NODE
- -savehostfile
- MP_SAVEHOSTFILE
- -spname
- SP_NAME
The following Job Specification flags override the associated environment
variables.
- -cmdfile
- MP_CMDFILE
- -newjob
- MP_NEWJOB
- -pgmmodel
- MP_PGMMODEL
The following I/O Control flags override the associated environment
variables.
- -labelio
- MP_LABELIO
- -stdinmode
- MP_STDINMODE
- -stdoutmode
- MP_STDOUTMODE
The following VT Trace Collection flags override the associated environment
variables.
- -samplefreq or -sfreq
- MP_SAMPLEFREQ
- -tbuffsize or -tbsize
- MP_TBUFFSIZE
- -tbuffwrap or -tbwrap
- MP_TBUFFWRAP
- -tmpdir
- MP_TMPDIR
- -tracedir or -tdir
- MP_TRACEDIR
- -tracefile or -tfile
- MP_TRACEFILE
- -tracelevel or -tlevel
- MP_TRACELEVEL
- -ttempsize or -ttsize
- MP_TTEMPSIZE
The following generation of diagnostic information flags override the
associated environment variables.
- -infolevel or -ilevel
- MP_INFOLEVEL
- -pmdlog
- MP_PMDLOG
The following Message Passing flags override the associated environment
variables.
- -buffer_mem
- MP_BUFFER_MEM
- -css_interrupt
- MP_CSS_INTERRUPT
- -eager_limit
- MP_EAGER_LIMIT
- -intrdelay
- MP_INTRDELAY
- -max_typedepth
- MP_MAX_TYPEDEPTH
- -use_flow_control
- MP_USE_FLOW_CONTROL
- -thread_stacksize
- MP_THREAD_STACKSIZE
- -single_thread
- MP_SINGLE_THREAD
- -wait_mode
- MP_WAIT_MODE
- -polling_interval
- MP_POLLING_INTERVAL
The following are miscellaneous flags:
- -euidevelop
- Overrides the MP_EUIDEVELOP environment variable.
- -pmlights
- Determines the number of lights displayed (per row) on the Program Marker
Array. This overrides the MP_PMLIGHTS environment
variable. For more information on the Program Marker Array, refer to
the manual page for the pmarray command.
- -usrport
- Overrides the MP_USRPORT environment variable.
- -coredir
- Overrides the MP_COREDIR environment variable.
DESCRIPTION
The poe command invokes the Parallel Operating Environment for
loading and executing programs on remote nodes. You can enter it at
your home node to:
- load and execute an SPMD program on all nodes of your partition.
- individually load the nodes of your partition with an MPMD job.
- load and execute a series of SPMD and MPMD programs, in individual job
steps, on the same partition.
- run non-parallel programs on remote nodes.
The operation of POE is influenced by a number of POE environment
variables. The flag options on this command are each used to
temporarily override one of these environment variables. User
program_options can be freely interspersed with the flag options, and
additional_options not to be parsed by POE can be placed after a
fence_string defined by the MP_FENCE environment
variable. If no program is specified, POE will either prompt
you for programs to load, or, if the MP_CMDFILE environment
variable is set, will load the partition using the specified commands
file.
The environment variables and flags that influence the operation of this
command fall into distinct categories of function. They are:
- Partition Manager control. The environment variables and
flags in this category determine the method of node allocation, message
passing mechanism, and the PULSE monitor function.
- Job specification. The environment variables and flags
in this category determine whether or not the Partition Manager should
maintain the partition for multiple job steps, whether commands should be read
from a file or STDIN, and how the partition should be loaded.
- I/O control. The environment variables and flags in this
category determine how I/O from the parallel tasks should be handled.
These environment variables and flags set the input and output modes, and
determine whether or not output is labeled by task id.
- VT trace collection. The environment variables and flags
in this category determine if and how execution traces are collected for
playback using the Visualization Tool (vt). They determine
which types of traces are collected, and how trace storage is handled.
- Generation of diagnostic information. The environment
variables and flags in this category enable you to generate diagnostic
information that may be required by the IBM Support Center in resolving
PE-related problems.
- Message Passing Interface. The environment variables and
flags in this category enable you to specify values for tuning message passing
applications.
- Miscellaneous. The additional environment variables and
flags in this category enable additional error checking, and set a dispatch
priority class for execution.
ENVIRONMENT VARIABLES
The environment variable descriptions in this section are grouped by
function.
The following environment variables are associated with Partition Manager
control.
- MP_ADAPTER_USE
- Determines how the node's adapter should be used. If using
LoadLeveler, the US communication subsystem library does not require dedicated
use of the SP switch on the node. Adapter use will be defaulted, as in Table 4, but shared usage may be specified. If using the Resource
Manager, this value is only used when POE is requesting non-specific nodes via
the MP_RMPOOL or -rmpool setting. Valid values are
dedicated and shared. If not set, the default is
dedicated for US jobs, or shared for IP jobs. The value of this
environment variable can be overridden using the -adapter_use
flag.
- MP_AUTH
- Determines the type of user authorization to be used. Valid values
are AIX (the default) for AIX based user authorization, using
/etc/hosts.equiv or .rhosts entries, and
DFS for DFS/DCE based authorization. This value can be
overridden with an entry in the /etc/poe.limits file.
There is no associated command line flag.
- MP_CPU_USE
- Determines how the node's CPU should be used. If using
LoadLeveler, the US communication subsystem library does not require unique
CPU use on the node. CPU use will be defaulted, as in Table 4, but multiple use may be specified. If using the Resource
Manager, this value is only used when POE is requesting non-specific nodes via
the MP_RMPOOL or -rmpool setting. Valid values are
multiple and unique. If not set, the default is
unique for US jobs, or multiple for IP jobs. The value of this
environment variable can be overridden using the -cpu_use
flag.
- MP_EUIDEVICE
- Determines the adapter set to use for message passing. Valid values
are en0 (for Ethernet), fi0 (for FDDI), tr0 (for
token-ring), and css0 (for the SP system's high performance
switch feature).
- MP_EUILIB
- Determines the communication subsystem library implementation to use for
communication - either the IP communication subsystem or the US
communication subsystem. In order to use the US communication
subsystem, you must have an SP system configured with its high performance
switch feature. Valid, case-sensitive, values are ip (for the
IP communication subsystem) or us (for the US communication
subsystem). The value of this environment variable can be overridden
using the -euilib flag.
- MP_EUILIBPATH
- Determines the path to the message passing and communication subsystem
libraries. This only needs to be set if the libraries are moved.
Valid values are any path specifier. The value of this environment
variable can be overridden using the -euilibpath flag.
- MP_HOSTFILE
- Determines the name of a host list file for node allocation. Valid
values are any file specifier. If not set, the default is
host.list in your current directory. The value of
this environment variable can be overridden using the -hostfile or
-hfile flags.
- MP_PROCS
- Determines the number of program tasks. Valid values are any number
from 1 to 128. If not set, the default is 1. The value of this
environment variable can be overridden using the -procs flag.
- MP_PULSE
- The interval (in seconds) at which POE checks the remote nodes to ensure
that they are communicating with the home node. The default interval is
600 seconds (10 minutes). To disable the pulse function, specify an
interval of 0 (zero) seconds. The pulse function is automatically
disabled when running the pdbx or pedb debuggers.
You can override the value of this environment variable with the
-pulse flag.
- MP_REMOTEDIR
- Specifies the name of a script which echoes the name of the current
directory to be used on the remote nodes. By default, the current
directory is the current directory at the time that POE is run. You may
need to specify this if the AutoMount Daemon is used to mount user file
systems, and the user is not using the Korn shell.
The script mpamddir is provided for mapping the C shell
directory name to an AutoMount Daemon name.
- MP_RESD
- Determines whether or not the Partition Manager should connect to LoadLeveler or the SP system Resource Manager to allocate nodes. Valid
values are either yes or no, and there is no default.
The value of this environment variable can be overridden using the
-resd flag.
- MP_RETRY
- Determines the period of time (in seconds) between processor node
allocation retries if there are not enough processor nodes immediately
available to run a program. This is only valid if you are using LoadLeveler or the SP system Resource Manager. Valid values are any integer
greater than or equal to 0. The default is 0 (no retry). The
value of this environment variable can be overridden using the -retry
flag.
- MP_RETRYCOUNT
- The number of times (at the interval set by MP_RETRY) that the
Partition Manager should attempt to allocate processor nodes. Valid
values are any integer greater than or equal to 0. If not set, the
default is 0. The value of this environment variable can be overridden
using the -retrycount flag.
- MP_MSG_API
- Indicates to POE which message passing API is being used by the parallel
tasks. You need to set this environment variable if a parallel task is
using LAPI alone or in conjunction with MPI. You do not need to set it
if a parallel task is using MPI only. The value of this environment
variable can be overridden using the -msg_api flag.
- MP_RMPOOL
- With regard to LoadLeveler, determines the name or number of the pool that
should be used for non-specific node allocation. With regard to the
Resource Manager, determines the number of the SP system pool that should be
used for non-specific node allocation. This environment
variable/command-line flag only applies to LoadLeveler or the SP system
Resource Manager. Valid values are any identifying pool name or number
for LoadLeveler, and any identifying pool number for the Resource
Manager. There is no default. The value of this environment
variable can be overridden using the -rmpool flag.
- MP_NODES
- Specifies the number of physical nodes on which to run the parallel
tasks. It may be used alone or in conjunction with
MP_TASKS_PER_NODE and/or MP_PROCS, as described in Table 7. The value of this environment variable can be
overridden using the -nodes flag.
- MP_TASKS_PER_NODE
- Specifies the number of tasks to be run on each of the physical
nodes. It may be used in conjunction with MP_NODES and/or
MP_PROCS, as described in Table 7, but may not be used alone. The value of this
environment variable can be overridden using the -tasks_per_node
flag.
- MP_SAVEHOSTFILE
- The name of an output host list file to be generated by the Partition
Manager. Valid values are any relative or full path name. The
value of this environment variable can be overridden using the
-savehostfile flag.
- MP_TIMEOUT
- Controls the length of time POE waits before abandoning an attempt to
connect to the remote nodes. The default is 150 seconds.
MP_TIMEOUT also changes the length of time the communication
subsystem will wait for a connection to be established during message passing
initialization.
- SP_NAME
- Determines the job management system (LoadLeveler or the SP system
Resource Manager) to use. If all nodes to be used for the parallel job
exist in a PSSP 2.4.0 partition, the SP_NAME
environment variable should be set to the name of the control workstation of
the SP system on which these nodes exist. This is the only case that
results in POE contacting the Resource Manager rather than LoadLeveler for
node allocation requests. When running POE from a workstation that is
external to the LoadLeveler cluster, the LoadL.so fileset must
be installed on the external node (see Using and Administering
LoadLeveler and IBM Parallel Environment for AIX: Installation for more information). When running POE from a workstation that is
external to the SP system, and using the Resource Manager, the
ssp.clients fileset must be installed on the external node
(see IBM Parallel Environment for AIX: Installation for more information).
- MP_CHECKFILE
- Defines the base name of the checkpoint file when checkpointing or
restarting a program. See "Checkpointing and Restarting Programs" for more information.
- MP_CHECKDIR
- Defines the directory where the checkpoint file will reside when
checkpointing or restarting a program. See "Checkpointing and Restarting Programs" for more information.
The following environment variables are associated with Job
Specification.
- MP_CMDFILE
- Determines the name of a POE commands file used to load the nodes of your
partition. If set, POE will read the commands file rather than
STDIN. Valid values are any file specifier. The value of this
environment variable can be overridden using the -cmdfile
flag.
- MP_NEWJOB
- Determines whether or not the Partition Manager maintains your partition
for multiple job steps. Valid values are yes or
no. If not set, the default is no. The value
of this environment variable can be overridden using the -newjob
flag.
- MP_PGMMODEL
- Determines the programming model you are using. Valid values are
spmd or mpmd. If not set, the default is
spmd. The value of this environment variable can be overridden
using the -pgmmodel flag.
The following environment variables are associated with I/O Control.
- MP_LABELIO
- Determines whether or not output from the parallel tasks are labeled by
task id. Valid values are yes or no. If not
set, the default is no. The value of this environment variable
can be overridden using the -labelio flag.
- MP_STDINMODE
- Determines the input mode - how STDIN is managed for the parallel
tasks. Valid values are:
- all
- all tasks receive the same input data from STDIN.
- none
- no tasks receive input data from STDIN; STDIN will be used by the home
node only.
- n
- STDIN is only sent to the task identified (n).
If not set, the default is all. The value of this
environment variable can be overridden using the -stdinmode
flag.
- MP_HOLD_STDIN
- Determines whether or not sending of STDIN from the home node to the
remote nodes is deferred until the message passing partition has been
established. Valid values are yes or no. If
not set, the default is no.
- MP_STDOUTMODE
- Determines the output mode - how STDOUT is handled by the parallel
tasks. Valid values are:
- unordered
- all tasks write output data to STDOUT asynchronously.
- ordered
- output data from each parallel task is written to its own buffer.
Later, all buffers are flushed, in task order, to STDOUT.
- a task id
- only the task indicated writes output data to STDOUT.
If not set, the default is unordered. The value of this
environment variable can be overridden using the -stdoutmode
flag.
The following environment variables are associated with VT Trace
Collection.
- MP_SAMPLEFREQ
- Determines the interval (in milliseconds) at which AIX kernel statistics
are sampled when executing a program with tracing on. Valid values are
any integer greater than or
equal to 0. If not set, the default is 10. The value of this
environment variable can be overridden using the -samplefreq
or -sfreq flags.
- MP_TBUFFSIZE
- Determines the size (in bytes) of the buffer used when generating trace
files. This may be specified as nnnK
or nnnM. If not set, the default is 5M. The value of
this environment variable can be overridden using the -tbuffsize
or -tbsize flags.
- MP_TBUFFWRAP
- Determines that a wraparound storage approach for trace records should be
used instead of the default three-tiered approach. With this approach,
the system keeps overwriting the buffer instead of flushing it to a temp
file. Valid values are either yes or no. If
not set, the default is no. The value of this environment
variable can be overridden using the -tbuffwrap
or -tbwrap flags.
- MP_TMPDIR
- The temporary directory to which output trace files are written.
Valid values are any path specifier. If not set, the default is
/tmp/username. The value of this environment variable can be
overridden using the -tmpdir flag.
- MP_TRACEDIR
- Determines the directory to which the final integrated trace file is
built. Valid values are any path specifier. If not set, the
default is the current directory. The value of this environment
variable can be overridden using the -tracedir
or -tdir flags.
- MP_TRACEFILE
- Determines the name of the output trace file created when executing a
program with tracing on. Valid values are any file specifier. If
not set, the default is the name of the program with the suffix
.trc added. The value of this environment variable
can be overridden using the -tracefile
and -tfile flags.
- MP_TRACELEVEL
- Determines the level of VT tracing that should be generated during the
execution of a program. Valid values are:
- 0
- no trace records
- 1
- Application Markers
- 2
- AIX Kernel Statistic and Application Markers
- 3
- Message Passing, Collective Communication, and Application Markers
- 9
- all trace records
If not set, the default is 0 (no trace records). The value
of this environment variable can be overridden using the -tracelevel
or -tlevel flags.
- MP_TTEMPSIZE
- Determines the size (in bytes) of the temp file used when generating trace
files. This may be specified as nnnM or
nnnG. If not set, the default is 10M. The value of
this environment variable can be overridden using the -ttempsize
or -ttsize flags.
The following environment variables are associated with the generation of
diagnostic information.
- MP_INFOLEVEL
- Determines the level of message reporting. Valid values are:
- 0
- error
- 1
- warning and error
- 2
- informational, warning, and error
- 3
- informational, warning, and error. Also reports diagnostic messages
for use by the IBM Support Center.
- 4, 5, 6
- Informational, warning, and error. Also reports high- and low-level
diagnostic messages for use by the IBM Support Center.
If not set, the default is 1 (warning and error). The value
of this environment variable can be overridden using the -infolevel
or -ilevel flags.
- MP_PMDLOG
- Determines whether or not diagnostic messages should be logged to a file
in /tmp on each of the remote nodes. Typically, this
environment variable/command-line flag is only used under the direction of the
IBM Support Center in resolving a PE-related problem. Valid values are
yes or no. If not set, the default is
no. The value of this environment variable can be overridden
using the -pmdlog flag.
- MP_DEBUG_LOG
- Determines the level of diagnostic messages written to
$MP_tmp/dbelog.pid.taskid. Typically, this
environment variable/command-line flag is only used under the direction of the
IBM Support Center in resolving a PE-related problem. This environment
variable has no associated command-line flag.
- MP_DEBUG_INITIAL_STOP
- Determines the initial breakpoint in the application where
pdbx or pedb will get control.
MP_DEBUG_INITIAL_STOP should be specified as
file_name:line_number. The line_number is
the number of the line within the source file file_name; where
file_name has been compiled with -g. The line
number has to be one that defines executable code. In general, this is
a line of code for which the compiler generates machine level code.
Another way to view this is that the line number is one for which debuggers
will accept a breakpoint. Another valid string for
MP_DEBUG_INITIAL_STOP would be the function_name of the
desired initial stopping point in the debugger. If this variable is not
specified, the default is to stop at the first executable source line in the
main routine. This environment variable has no associated command-line
flag.
- MP_PMDSUFFIX
-
When using LoadLeveler, this environment variable determines a string to be
appended to the normal partition manager daemon executable. The normal
partition manager daemon executable specified is /etc/pmdv2.
By setting MP_PMDSUFFIX, you can append a string to
pmdv2. If MP_PMDSUFFIX is set to abc, for
example, then the partition manager daemon that gets run on each node of the
parallel task is /etc/pmdv2abc. When using the Resource
Manager, this environment variable determines a string to be appended to the
normal tcp service. The normal tcp service specified in
/etc/services is named pmv2. By setting
MP_PMDSUFFIX, you can append a string to pmv2. If
MP_PMDSUFFIX is set to abc, for example, then the service
requested in /etc/services is pmv2abc. Using the
environment variable with LoadLeveler or the Resource Manager as described
above permits testing of alternate versions of the Partition Manager
daemon. Typically, this environment variable is only used under the
direction of the IBM Support Center in resolving a PE-related problem.
Valid values are any string. This environment variable has no
associated command-line flag.
The following environment variables are associated with the Message Passing
Interface.
- MP_BUFFER_MEM
- Changes the maximum size of memory used by the communication subsystem to
buffer early arrivals. The default is 2.8 megabytes for IP and 64 megabytes for
US. However, if checkpointing a program, for US the default will be
2.8 megabytes. If you are using this environment variable to
change the maximum size of memory used by the communication subsystem while
checkpointing a program, please be aware that the amount of space needed for
the checkpointing files will be increased by the value of
MP_BUFFER_MEM.
- MP_CSS_INTERRUPT
- Determines whether or not arriving message packets cause
interrupts. This may provide better performance for certain
applications. Valid values are yes and no.
If not set, the default is no.
- MP_EAGER_LIMIT
- Changes the threshold value for message size, above which rendezvous
protocol is used.
- MP_INTRDELAY
- Allows user programs to tune the delay parameter without having to
recompile existing applications.
- MP_MAX_TYPEDEPTH
- Changes the maximum depth of message buffer types.
- MP_USE_FLOW_CONTROL
- Limits the maximum number of outstanding messages posted by a
sender.
- MP_THREAD_STACKSIZE
- Determines the additional stacksize allocated for user programs executing
on an MPI service thread. If you allocate insufficient space, the
program may encounter a SIGSEGV exception.
- MP_SINGLE_THREAD
- Avoids mutex lock overheads in a single threaded
program. This is an optimization flag, with values of no and
yes. The default value is no, which means
multiple user message passing threads are assumed.
| Note: | MPI-IO cannot be used if this is set to YES. Results are
undefined if this is YES, with multiple message passing threads in
use.
|
- MP_WAIT_MODE
- Determines how a thread or task behaves when it discovers it is blocked,
waiting for a message to arrive. Values are poll,
yield, and sleep. The default mode for the signal
handling library is poll for US, and sleep for IP.
- MP_POLLING_INTERVAL
- Defines the polling interval in microseconds. The maximum interval
is approximately 2 billion microseconds (2000 seconds). The default is
180,000 microseconds for IP, and 400,000 microseconds for US.
The following are miscellaneous environment variables:
- MP_EUIDEVELOP
- Determines whether or not the message passing interface performs more
detailed checking during execution. This additional checking is
intended for developing applications, and can significantly slow
performance. Valid values are yes or no, deb
(for "debug"), nor (for "normal"), and min (for
"minimum"). The debug and min values are used to start and stop parameter checking. If not
set, the default is no. The value of this environment
variable can be overridden using the -euidevelop flag.
- MP_FENCE
- Determines a fence_string to be used for separating options you
want parsed by POE from those you do not. Valid values are any string,
and there is no default. Once set, you can then use the
fence_string followed by additional_options on the
poe command line. The additional_options will not be
parsed by POE. This environment variable has no associated command-line
flag.
- MP_NOARGLIST
- Determines whether or not POE ignores the argument list. Valid
values are yes and no. If set to yes, POE
will not attempt to remove POE command-line flags before passing the argument
list to the user's program. This environment variable has no
associated command-line flag.
- MP_PMLIGHTS
- Indicates the number of lights displayed per row on the Program Marker
Array.
- MP_PRIORITY
- Determines a dispatch priority adjustment class for execution. See
IBM Parallel Environment for AIX: Installation for more information on dispatch priority classes. Valid values are
any of the dispatch priority classes set up by the system administrator in the
file /etc/poe.priority. This environment variable has
no associated command-line flag.
- MP_USRPORT
- Indicates the port id used by the Partition Manager to connect to the
Program Marker Array. By default, the Partition Manager connects to the
Array using a socket assigned to port 9999. If you get an error message
indicating that the port is in use, specify a different port. Standard
TCP/IP practice suggests using ports greater than 5000 and less than
10000.
- MP_COREDIR
- Creates a separate directory for each task's core file.
EXAMPLES
- Assume the MP_PGMMODEL environment variable is set to
spmd, and MP_PROCS is set to 6. To load
and execute the SPMD program sample on the six remote nodes of your
partition, enter:
poe sample
- Assume you have an MPMD application consisting of two programs -
master and workers. These programs are designed to
run together and communicate via calls to message passing subroutines.
The program master is designed to run on one processor node.
The workers program is designed to run as separate tasks on any
number of other nodes. The MP_PGMMODEL environment variable
is set to mpmd, and MP_PROCS is set to 6.
To individually load the six remote nodes with your MPMD application,
enter:
poe
Once the partition is established, the poe command responds with
the prompt:
0:host1_name>
To load the master program as task 0 on host1_name, enter:
master
The poe command responds with a prompt for the next node to
load. When you have loaded the last node of your partition, the
poe command displays the message Partition
loaded... and begins execution.
- Assume you want to run three SPMD programs - setup,
computation, and cleanup - as job steps on the same
partition of nodes. The MP_PGMMODEL environment variable is
set to spmd, and MP_NEWJOB is set to yes.
You enter:
poe
Once the partition is established, the poe command responds with
the prompt:
Enter program name (or quit):
To load the program setup, enter:
setup
The program setup executes on all nodes of your partition. When
execution completes, the poe command again prompts you for a program
name. Enter the program names in turn. To release the partition,
enter:
quit
- To check the process status (using the non-parallel command ps)
for all remote nodes in your partition, enter:
poe ps
FILES
host.list (Default host list file)
RELATED INFORMATION
Commands: mpcc(1), mpcc_r(1), mpCC(1),
mpCC_r(1), mpxlf(1), mpxlf_r(1), pdbx(1),
pedb(1), pmarray(1), vt(1), xprofiler(1)
NAME
poeauth - Allows you to copy Distributed File System (DFS)
credentials to all nodes on which you want to run POE jobs.
SYNOPSIS
poeauth [POE options]
DESCRIPTION
The poeauth command allows you to copy DFS credentials to all
nodes on which you want to run POE jobs. You can use any POE command
line flag or environment variable with poeauth, because it is a POE
application. Before running poeauth, you must run
dce_login from task 0 (where DFS/DCE credentials reside).
The credentials must reside on task 0 for poeauth to copy
them. In order to be properly authorized, you must run
poeauth before running any POE applications. When the
credentials are copied, there is no need to use poeauth until the
credentials expire.
Return codes are:
| Note: | The actual command return code value is 128 plus the unsigned return code
value. That is, a return code of -2 will give a value of 130.
For more information, see the "Exit Status" section in IBM Parallel
Environment for AIX: MPI Programming and Subroutine Reference
|
- -1
- error reading KRB5CCNAME environment variable
- -3
- credentials files not found on home node, task 0
- -4
- could not open credentials file for read
- -5
- no room on destination node's filesystem
- -6
- error opening file output file
- -7
- error creating output file
- -8
- error writing to output file
- -9
- MPI_Send of data failed
- -10
- Final MPI_Send failed
- -11
- MPI_Recv failed
- -13
- error creating unique id for credentials
- -14
- error opening /tmp/poedce_master file
- -15
- error creating /tmp/poedce_master file
- -16
- error writing to /tmp/poedce_master file
EXAMPLES
- To copy the credentials to the first 32 nodes listed in the host list file
named "my_hosts" in your home directory, enter:
poeauth -procs 32 -hfile $HOME/my_hosts -labelio yes -pmdlog yes
| Note: | If you were previously set up to use a PMD suffix, remember to set the
MP_PMDSUFFIX environment variable and export it before running
poeauth.
|
RELATED INFORMATION
Commands: poe(1), pdbx(1), pedb(1)
NAME
poekill - terminates all remote tasks for a give
program.
SYNOPSIS
poe poekill pgm_name [poe_options]
or
rsh remote_node poekill pgm_name
poekill is a Korn shell script that searches for the existence
of running programs (named pgm_name) owned by the user, and
terminates them via SIGTERM signals. If run under POE,
poekill uses the standard POE mechanism for identifying the set of
remote nodes (host.list, Resource Manager, etc.). If run
under rsh, poekill applies only to the node specified as
remote_node.
FLAGS
When run as a POE program, standard POE flags apply.
DESCRIPTION
poekill determines the user ID of the user that submitted the
command. It then uses the ID to obtain a list of active processes,
which is filtered by the pgm_name argument into a scratch file in
/tmp. The file is processed by an awk script that sends a SIGTERM
signal (15) to each process in the list, and echoes the action back to the
user. The scratch file is then erased, and the script exits with code
of 0.
If you do not provide a pgm_name, an error message is printed and the
script exits with a code of 0.
The pgm_name can be a substring of the program name.
RELATED INFORMATION
Commands: rsh(1), poe(1), kill(1)
NAME
poestat - Starts the System Status Array, which is an
X-Windows tool for monitoring the operational status and CPU utilization of
processor nodes.
SYNOPSIS
poestat [-norm]
The poestat command starts the System Status Array. This
X-Windows tool lets you quickly survey the utilization of processor
nodes. Normally, this would be run in the background.
FLAGS
- -norm
- Indicates that a job management system, which would normally be used to select nodes for monitoring, is
not available. Instead, the System Status Array program selects for
monitoring each of the nodes on the LAN that have the VT Statistics Collector
Daemon (digd) running. It also selects each of the nodes
listed in the host list file indicated by the MP_HOSTFILE
environment variable. If you are monitoring nodes of an RS/6000 network
cluster, you must use this flag.
DESCRIPTION
The poestat command starts the System Status Array. This
X-Windows monitoring tool lets you quickly survey the utilization of processor
nodes. The Array consists of a number of squares, each representing a
processor node of your SP system or cluster. The squares are colored
pink and yellow to show the instantaneous percent of CPU utilization for each
processor node. If a square were to appear all pink, the node would be
at 0 percent utilization. If a square were to appear all yellow, it
would be at 100 percent utilization. To the right of the Array is a
node list which contains the name of each node shown in the Array. The
nodes are listed in the order in which they were contacted, left to right,
starting with the top row of the Array. You use this list to identify
the name of a node represented in the Array.
In order to use this tool, the Visualization Tool's Statistics
Collector Daemon process (digd) needs to be running on each of the
nodes you wish to monitor. The daemon feeds the System Status Array
with the CPU information it displays, and is created on each of your nodes as
part of the Visualization Tool's installation procedure. The
digd statistics collector daemon can also feed information to the
Visualization Tool. If a square on the Array appears gray, the node is
unavailable for monitoring. It either does not have the Statistics
Collector daemon running, or the Array cannot communicate with it.
ENVIRONMENT VARIABLES
- MP_HOSTFILE
- This environment variable is normally associated with node
allocation. However, it is also checked by the poestat command
when running with the -norm option. It determines the name of
a host list file to use in selecting nodes for monitoring. If not set,
the default is host.list in your current directory.
- MP_RESD
- Identifies whether or not node data for allocated nodes received from a
job management system (LoadLeveler or the SP system Resource Manager) should
be filtered with a host list file. If MP_RESD=yes, and a host
list file is specified, only those nodes allocated by the job management
system and listed in the host list file will be displayed. This option
is useful if you would like to view selected job management system allocated
nodes or view up to 512 nodes. To view 512 nodes, create two host list
files. Each file can contain up to 255 nodes. Set
MP_HOSTFILE to one of the host list files and start poestat
in the background. Set MP_HOSTFILE to the other host list file
and start another poestat in the background. Two instances of
poestat will be running, each displaying a different set of 255
nodes. If MP_RESD=no, the job management system node data will
not be displayed (similar to the -norm command line option).
For more information, see "Step 3e: Set the MP_RESD Environment Variable".
- SP_NAME
- Identifies the control workstation of the SP system. This should
only be used when all nodes to be monitored exist in a PSSP
2.3.0 or 2.4.0 partition. This is the only
case that results in poestat contacting the Resource Manager rather
than LoadLeveler for node allocation requests. When running
poestat from a workstation that is external to the SP system, and
using the Resource Manager, the ssp.clients fileset must be
installed on the external node (see IBM Parallel Environment for
AIX: Installation for more information).
EXAMPLES
To start the System Status Array as a background process when the SP system
Resource Manager is available, enter:
poestat &
To start the System Status Array as a background process to monitor an
RS/6000 network cluster, enter:
poestat -norm &
To display 512 resource manager nodes, enter:
export MP_RESD=yes
export MP_HOSTFILE= a host list file with 1st 255 nodes
poestat &
export MP_HOSTFILE= a host list file with 2nd 255 nodes
poestat &
FILES
host.list (Default host list file)
RELATED INFORMATION
Commands: vt(1)
[ Top of Page | Previous Page | Next Page | Table of Contents | Index ]