No Hw-module Slot 1 Oversubscription

This week I ran into an oversubcription issue on an ASA5550. To alleviate the issue, we followed the recommendations below from Cisco. I am including some of the conditions I saw before the change. Keyword is Alleviate, depending on your traffic rates you might resolve the problem going this route. In other cases, you would just have to get a second pair or firewalls to segregatetraffic or just upgrade to 10GB. The best way to determine this is to place a sniffer between the ASA and drill down as close to the microsecond to see the microbursts on the line and data rate patterns.

• No Hw-module Slot 1 Oversubscription Port-group 1
• No Hw-module Switch 1 Slot 1 Oversubscription Port-group 1

No Hw Module Switch 1 Slot 1 Oversubscription, china moon pennyslots slotomania, massage during poker, 4 pics 1 word owl panther roulette man. Lists each dataplane on the installed NPCs. The first two characters indicate the slot number and the last three characters indicate the dataplane number. For example, s1dp0 indicates dataplane 0 on the NPC in slot 1 and s1dp1 indicates dataplane 1 on the NPC in slot1.

Maximizing Throughput (ASA 5550)

This is a very popular no deposit bonus that offers you a huge sum of money as bonus – usually between $500 and $3000. No Hw Module Switch 1 Slot 1 Oversubscription You are required to use up the entire bonus money in a given time, usually 60 minutes. Any bonus money left over after the time period has expired becomes unusable.

Slot-1 Stack # download image 1.1.1.1 summitX-16.1.1.4.xos slot 2 secondary Debug information files are present in internal-memory. These files will be removed if you continue with download. SRX5400, SRX5600, and SRX5800 Services Gateway Card Overview, Cards Supported on SRX5400, SRX5600, and SRX5800 Services Gateways, SRX5400 Services Gateway Card Cage and Slots, SRX5400 Services Gateway Services Processing Card Overview, Services Processing Card SRX5K-SPC-2-10-40 Specifications, Services Processing Card SRX5K-SPC-4-15-320 Specifications, Services Processing Card SRX5K.

Show Traffic
----------------------------------------
Per Slot Throughput Profile (1 minute)
----------------------------------------
Packets-per-second profile:
Slot 0: 12654 89% ********************************************
Slot 1: 1603 11% *****
Bytes-per-second profile:
Slot 0: 1649003 76% **************************************
Slot 1: 511183 24% ************
On the interface level, you would see the Underruns counter increment along with the Overruns counter (See below). To try and alleviate or resolve this issue move one of the ports to Gi1/X and mmonitor it over a few days.
Per Cisco:

ASA5550/act# show interface gigabitEthernet 0/1
Interface GigabitEthernet0/1 'HM', is up, line protocol is up
Hardware is i82546GB rev03, BW 1000 Mbps, DLY 10 usec
Auto-Duplex(Full-duplex), Auto-Speed(1000 Mbps)
Input flow control is unsupported, output flow control is off
Description: 6509
MAC address 6400.f182.6771, MTU 1500
IP address 192.168.1.1 subnet mask 255.255.255.0
24794625 packets input, 4336231091 bytes, 0 no buffer
Received 4648 broadcasts, 0 runts, 0 giants
0 input errors, 0 CRC, 0 frame, 0 overrun, 0 ignored, 0 abort
0 pause input, 0 resume input
0 L2 decode drops
40981082 packets output, 3012528711 bytes, 1614642 underruns
0 pause output, 0 resume output
0 output errors, 0 collisions, 0 interface resets
0 late collisions, 0 deferred
0 input reset drops, 0 output reset drops, 0 tx hangs
input queue (blocks free curr/low): hardware (255/230)
output queue (blocks free curr/low): hardware (255/0)
Traffic Statistics for 'HM':
23737668 packets input, 3724976676 bytes
42595724 packets output, 2342955016 bytes
6597 packets droppedTable of Contents

Name

orterun, mpirun, mpiexec - Execute serial and paralleljobs in Open MPI. oshrun, shmemrun - Execute serial and parallel jobs inOpen SHMEM.

Note:mpirun, mpiexec, and orterun are all synonyms for eachother as well as oshrun, shmemrun in case Open SHMEM is installed. Usingany of the names will produce the same behavior.

Synopsis

Single ProcessMultiple Data (SPMD) Model:

mpirun [ options ] <program> [ <args> ]

MultipleInstruction Multiple Data (MIMD) Model:

mpirun [ global_options ] [ local_options1 ]
<program1> [ <args1> ] : [ local_options2 ]
<program2> [ <args2> ] : ... :
[ local_optionsN ]
<programN> [ <argsN> ]

Note that in both models, invoking mpirun via an absolutepath name is equivalent to specifying the --prefix option with a <dir> valueequivalent to the directory where mpirun resides, minus its last subdirectory. For example:

% /usr/local/bin/mpirun ...

is equivalent to

% mpirun --prefix /usr/local

Quick Summary

If you are simply looking for how to run an MPI application,you probably want to use a command line of the following form:

% mpirun[ -np X ] [ --hostfile <filename> ] <program>

This will run X copies of <program> in your current run-time environment(if running under a supported resource manager, Open MPI’s mpirun will usuallyautomatically use the corresponding resource manager process starter, asopposed to, for example, rsh or ssh, which require the use of a hostfile,or will default to running all X copies on the localhost), scheduling (bydefault) in a round-robin fashion by CPU slot. See the rest of this pagefor more details.

Please note that mpirun automatically binds processesas of the start of the v1.8 series. Three binding patterns are used in theabsence of any further directives:

Bind to core:

when the number of processesis <= 2

Bind to socket:

when the number of processes is > 2

Bind to none:

when oversubscribed

If your application uses threads, then you probablywant to ensure that you are either not bound at all (by specifying --bind-tonone), or bound to multiple cores using an appropriate binding level orspecific number of processing elements per application process.

Options

mpirun will send the name of the directory where it was invoked on thelocal node to each of the remote nodes, and attempt to change to that directory. See the 'Current Working Directory' section below for further details.

<program>

The program executable. This is identified as the first non-recognizedargument to mpirun.

<args>

Pass these run-time arguments to every new process. These must always be the last arguments to mpirun. If an app context fileis used, <args> will be ignored.

-h, --help

Display help for this command

-q, --quiet

Suppress informative messages from orterun during applicationexecution.

-v, --verbose

Be verbose

-V, --version

Print version number. Ifno other arguments are given, this will also cause orterun to exit.

-N<num>

Launch num processes per node on all allocated nodes (synonym for npernode).

-display-map, --display-map

Display a table showing the mapped location ofeach process prior to launch.

-display-allocation, --display-allocation

Displaythe detected resource allocation.

-output-proctable, --output-proctable

Outputthe debugger proctable after launch.

-dvm, --dvm

Create a persistent distributedvirtual machine (DVM).

-max-vm-size, --max-vm-size <size>

Number of processesto run.

-novm, --novm

Execute without creating an allocation-spanning virtualmachine (only start daemons on nodes hosting application procs).

-hnp,--hnp <arg0>

Specify the URI of the Head Node Process (HNP), or the name ofthe file (specified as file:filename) that contains that info.

Use oneof the following options to specify which hosts (nodes) of the clusterto run on. Note that as of the start of the v1.8 release, mpirun will launcha daemon onto each host in the allocation (as modified by the followingoptions) at the very beginning of execution, regardless of whether or notapplication processes will eventually be mapped to execute there. This isdone to allow collection of hardware topology information from the remotenodes, thus allowing us to map processes against known topology. However,it is a change from the behavior in prior releases where daemons were onlylaunched after mapping was complete, and thus only occurred on nodes whereapplication processes would actually be executing.

-H, -host, --host <host1,host2,...,hostN>

List of hosts on which to invoke processes.

-hostfile, --hostfile <hostfile>

Provide a hostfile to use.

-default-hostfile, --default-hostfile <hostfile>

Provide a default hostfile.

-machinefile, --machinefile <machinefile>

Synonymfor -hostfile.

-cpu-set, --cpu-set <list>

Restrict launched processes to thespecified logical cpus on each node (comma-separated list). Note that thebinding options will still apply within the specified envelope - e.g., youcan elect to bind each process to only one cpu within the specified cpuset.

The following options specify the number of processes to launch.Note that none of the options imply a particular binding policy - e.g., requestingN processes for each socket does not imply that the processes will be boundto the socket.

-c, -n, --n, -np <#>

Run this many copies of the program on thegiven nodes. This option indicates that the specified file is an executableprogram and not an application context. If no value is provided for thenumber of copies to execute (i.e., neither the '-np' nor its synonyms areprovided on the command line), Open MPI will automatically execute a copyof the program on each process slot (see below for description of a 'processslot'). This feature, however, can only be used in the SPMD model and willreturn an error (without beginning execution of the application) otherwise. -<>

Launch N times the number of objects of the specified type on each node.

-npersocket, --npersocket <#persocket>

On each node, launch this many processestimes the number of processor sockets on the node. The -npersocket optionalso turns on the -bind-to-socket option. (deprecated in favor of --map-by ppr:n:socket)

-npernode, --npernode <#pernode>

On each node, launch this many processes.(deprecated in favor of --map-by ppr:n:node)

-pernode, --pernode

On each node,launch one process -- equivalent to -npernode 1. (deprecated in favor of --map-byppr:1:node)

To map processes:

--map-by <foo>

Map to the specified object,defaults to socket. Supported options include slot, hwthread, core, L1cache,L2cache, L3cache, socket, numa, board, node, sequential, distance, andppr. Any object can include modifiers by adding a : and any combinationof PE=n (bind n processing elements to each proc), SPAN (load balance theprocesses across the allocation), OVERSUBSCRIBE (allow more processes ona node than processing elements), and NOOVERSUBSCRIBE. This includes PPR,where the pattern would be terminated by another colon to separate it fromthe modifiers.

-bycore, --bycore

Map processes by core (deprecated in favorof --map-by core)

-byslot, --byslot

Map and rank processes round-robin by slot.

-nolocal, --nolocal

Do not run any copies of the launched application onthe same node as orterun is running. This option will override listingthe localhost with --host or any other host-specifying mechanism.

-nooversubscribe,--nooversubscribe

Do not oversubscribe any nodes; error (without startingany processes) if the requested number of processes would cause oversubscription.This option implicitly sets 'max_slots' equal to the 'slots' value foreach node. (Enabled by default).

-oversubscribe, --oversubscribe

Nodes areallowed to be oversubscribed, even on a managed system, and overloadingof processing elements.

-bynode, --bynode

Launch processes one per node, cyclingby node in a round-robin fashion. This spreads processes evenly among nodesand assigns MPI_COMM_WORLD ranks in a round-robin, 'by node' manner.

-cpu-list,--cpu-list <cpus>

Comma-delimited list of processor IDs to which to bind processes[default=NULL]. Processor IDs are interpreted as hwloc logical core IDs. Run the hwloc lstopo(1) command to see a list of available cores and theirlogical IDs.

To order processes’ ranks in MPI_COMM_WORLD:

--rank-by <foo>

Rank in round-robin fashion according to the specified object, defaultsto slot. Supported options include slot, hwthread, core, L1cache, L2cache,L3cache, socket, numa, board, and node.

For process binding:

--bind-to<foo>

Bind processes to the specified object, defaults to core. Supportedoptions include slot, hwthread, core, l1cache, l2cache, l3cache, socket,numa, board, cpu-list, and none.

-cpus-per-proc, --cpus-per-proc <#perproc>

Bindeach process to the specified number of cpus. (deprecated in favor of --map-by<obj>:PE=n)

-cpus-per-rank, --cpus-per-rank <#perrank>

Alias for -cpus-per-proc. (deprecatedin favor of --map-by <obj>:PE=n)

-bind-to-core, --bind-to-core

Bind processes to cores(deprecated in favor of --bind-to core)

-bind-to-socket, --bind-to-socket

Bind processesto processor sockets (deprecated in favor of --bind-to socket)

-report-bindings,--report-bindings

Report any bindings for launched processes.

For rankfiles:

-rf, --rankfile <rankfile>

Provide a rankfile file.

To manage standardI/O:

-output-filename, --output-filename <filename>

Redirect the stdout, stderr,and stddiag of all processes to a process-unique version of the specifiedfilename. Any directories in the filename will automatically be created.Each output file will consist of filename.id, where the id will be the processes’rank in MPI_COMM_WORLD, left-filled with zero’s for correct ordering in listings.A relative path value will be converted to an absolute path based on thecwd where mpirun is executed. Note that this will not work on environmentswhere the file system on compute nodes differs from that where mpirun isexecuted.

-stdin, --stdin <rank>

The MPI_COMM_WORLD rank of the process thatis to receive stdin. The default is to forward stdin to MPI_COMM_WORLD rank0, but this option can be used to forward stdin to any process. It is alsoacceptable to specify none, indicating that no processes are to receivestdin.

-merge-stderr-to-stdout, --merge-stderr-to-stdout

Merge stderr to stdoutfor each process.

-tag-output, --tag-output

Tag each line of output to stdout,stderr, and stddiag with [jobid, MCW_rank]<stdxxx> indicating the processjobid and MPI_COMM_WORLD rank of the process that generated the output,and the channel which generated it.

-timestamp-output, --timestamp-output

Timestampeach line of output to stdout, stderr, and stddiag.

-xml, --xml

Provide alloutput to stdout, stderr, and stddiag in an xml format.

-xml-file, --xml-file<filename>

Provide all output in XML format to the specified file.

-xterm,--xterm <ranks>

Display the output from the processes identified by their MPI_COMM_WORLDranks in separate xterm windows. The ranks are specified as a comma-separatedlist of ranges, with a -1 indicating all. A separate window will be createdfor each specified process. Note: xterm will normally terminate the windowupon termination of the process running within it. However, by adding a'!' to the end of the list of specified ranks, the proper options willbe provided to ensure that xterm keeps the window open after the processterminates, thus allowing you to see the process’ output. Each xterm windowwill subsequently need to be manually closed. Note: In some environments,xterm may require that the executable be in the user’s path, or be specifiedin absolute or relative terms. Thus, it may be necessary to specify a localexecutable as './foo' instead of just 'foo'. If xterm fails to find the executable,mpirun will hang, but still respond correctly to a ctrl-c. If this happens,please check that the executable is being specified correctly and try again.

To manage files and runtime environment:

-path, --path <path>

<path> thatwill be used when attempting to locate the requested executables. Thisis used prior to using the local PATH setting.

--prefix <dir>

Prefix directorythat will be used to set the PATH and LD_LIBRARY_PATH on the remote nodebefore invoking Open MPI or the target process. See the 'Remote Execution'section, below.

--noprefix

Disable the automatic --prefix behavior

-s, --preload-binary

Copy the specified executable(s) to remote machines prior to starting remoteprocesses. The executables will be copied to the Open MPI session directoryand will be deleted upon completion of the job.

--preload-files <files>

Preloadthe comma separated list of files to the current working directory of theremote machines where processes will be launched prior to starting thoseprocesses.

-set-cwd-to-session-dir, --set-cwd-to-session-dir

Set the working directoryof the started processes to their session directory.

-wd <dir>

Synonym for-wdir.

-wdir <dir>

Change to the directory <dir> before the user’s program executes.See the 'Current Working Directory' section for notes on relative paths.Note: If the -wdir option appears both on the command line and in an applicationcontext, the context will take precedence over the command line. Thus, ifthe path to the desired wdir is different on the backend nodes, then itmust be specified as an absolute path that is correct for the backend node.

-x <env>

Export the specified environment variables to the remote nodesbefore executing the program. Only one environment variable can be specifiedper -x option. Existing environment variables can be specified or new variablenames specified with corresponding values. For example: % mpirun -xDISPLAY -x OFILE=/tmp/out ...

The parser for the -x option is not very sophisticated; it does not evenunderstand quoted values. Users are advised to set variables in the environment,and then use -x to export (not define) them.

Setting MCA parameters:

-gmca, --gmca <key> <value>

Pass global MCA parameters that are applicable toall contexts. <key> is the parameter name; <value> is the parameter value.

-mca, --mca <key> <value>

Send arguments to various MCA modules. See the 'MCA'section, below.

-am <arg0>

Aggregate MCA parameter set file list.

-tune,--tune <tune_file>

Specify a tune file to set arguments for various MCA modulesand environment variables. See the 'Setting MCA parameters and environmentvariables from file' section, below.

For debugging:

-debug, --debug

Invokethe user-level debugger indicated by the orte_base_user_debugger MCA parameter.

--get-stack-traces

When paired with the --timeout option, mpirun will obtainand print out stack traces from all launched processes that are still alivewhen the timeout expires. Note that obtaining stack traces can take a littletime and produce a lot of output, especially for large process-count jobs.

-debugger, --debugger <args>

Sequence of debuggers to search for when --debugis used (i.e. a synonym for orte_base_user_debugger MCA parameter).

--timeout<seconds>

The maximum number of seconds that mpirun (also known as mpiexec,oshrun, orterun, etc.) will run. After this many seconds, mpirun will abortthe launched job and exit with a non-zero exit status. Using --timeout canbe also useful when combined with the --get-stack-traces option.

-tv, --tv

Launchprocesses under the TotalView debugger. Deprecated backwards compatibilityflag. Synonym for --debug.

There are also other options:

--allow-run-as-root

Allow mpirun to run when executed by the root user (mpirun defaults toaborting when launched as the root user). Be sure to see the Running asroot section, below, for more detail.

--app <appfile>

Provide an appfile,ignoring all other command line options.

-cf, --cartofile <cartofile>

Providea cartography file.

-continuous, --continuous

Job is to run until explicitlyterminated.

-disable-recovery, --disable-recovery

Disable recovery (resetsall recovery options to off).

-do-not-launch, --do-not-launch

Perform all necessaryoperations to prepare to launch the application, but do not actually launchit.

-do-not-resolve, --do-not-resolve

Do not attempt to resolve interfaces.

-enable-recovery, --enable-recovery

Enable recovery from process failure [Default= disabled].

-index-argv-by-rank, --index-argv-by-rank

Uniquely index argv[0] foreach process using its rank.

-leave-session-attached, --leave-session-attached

Do not detach OmpiRTE daemons used by this application. This allows errormessages from the daemons as well as the underlying environment (e.g., whenfailing to launch a daemon) to be output.

-max-restarts, --max-restarts <num>

Max number of times to restart a failed process.

-ompi-server, --ompi-server<uri or file>

Specify the URI of the Open MPI server (or the mpirun to beused as the server), the name of the file (specified as file:filename)that contains that info, or the PID (specified as pid:#) of the mpirunto be used as the server. The Open MPI server is used to support multi-applicationdata exchange via the MPI-2 MPI_Publish_name and MPI_Lookup_name functions.

-personality, --personality <list>

Comma-separated list of programming model,languages, and containers being used (default='ompi').

--ppr <list>

Comma-separatedlist of number of processes on a given resource type [default: none].

-report-child-jobs-separately, --report-child-jobs-separately

Return the exit statusof the primary job only.

-report-events, --report-events <URI>

Report eventsto a tool listening at the specified URI.

-report-pid, --report-pid <channel>

Print out mpirun’s PID during startup. The channel must be either a ’-’ to indicatethat the pid is to be output to stdout, a ’+’ to indicate that the pid isto be output to stderr, or a filename to which the pid is to be written.

-report-uri, --report-uri <channel>

Print out mpirun’s URI during startup. Thechannel must be either a ’-’ to indicate that the URI is to be output to stdout,a ’+’ to indicate that the URI is to be output to stderr, or a filename towhich the URI is to be written.

-show-progress, --show-progress

Output a briefperiodic report on launch progress.

-terminate, --terminate

Terminate theDVM.

-use-hwthread-cpus, --use-hwthread-cpus

Use hardware threads as independentcpus.

-use-regexp, --use-regexp

Use regular expressions for launch.

Thefollowing options are useful for developers; they are not generally usefulto most ORTE and/or MPI users:

-d, --debug-devel

Enable debugging of the OmpiRTE(the run-time layer in Open MPI). This is not generally useful for most users.

--debug-daemons

Enable debugging of any OmpiRTE daemons used by this application.

--debug-daemons-file

Enable debugging of any OmpiRTE daemons used by thisapplication, storing output in files.

-display-devel-allocation, --display-devel-allocation

Display a detailed list of the allocation being used by this job.

-display-devel-map,--display-devel-map

Display a more detailed table showing the mapped locationof each process prior to launch.

-display-diffable-map, --display-diffable-map

Display a diffable process map just before launch.

-display-topo, --display-topo

Display the topology as part of the process map just before launch.

-launch-agent,--launch-agent

Name of the executable that is to be used to start processeson the remote nodes. The default is 'orted'. This option can be used to testnew daemon concepts, or to pass options back to the daemons without havingmpirun itself see them. For example, specifying a launch agent of orted-mca odls_base_verbose 5 allows the developer to ask the orted for debuggingoutput without clutter from mpirun itself.

--report-state-on-timeout

When pairedwith the --timeout command line option, report the run-time subsystem stateof each process when the timeout expires.

There may be other options listedwith mpirun --help.

Environment Variables

MPIEXEC_TIMEOUT

Synonym for the--timeout command line option.

Description

One invocation of mpirun startsan MPI application running under Open MPI. If the application is singleprocess multiple data (SPMD), the application can be specified on the mpiruncommand line.

If the application is multiple instruction multiple data(MIMD), comprising of multiple programs, the set of programs and argumentcan be specified in one of two ways: Extended Command Line Arguments, andApplication Context.

An application context describes the MIMD program setincluding all arguments in a separate file. This file essentially containsmultiple mpirun command lines, less the command name itself. The abilityto specify different options for different instantiations of a programis another reason to use an application context.

Extended command line argumentsallow for the description of the application layout on the command lineusing colons (:) to separate the specification of programs and arguments.Some options are globally set across all specified programs (e.g. --hostfile),while others are specific to a single program (e.g. -np).

Specifying HostNodes

Host nodes can be identified on the mpirun command line with the-host option or in a hostfile.

For example,

mpirun -H aa,aa,bb ./a.out

launchestwo processes on node aa and one on bb.

Or, consider the hostfile

% cat myhostfile
aa slots=2
bb slots=2
cc slots=2

Here, we list both the host names (aa, bb, and cc) but also how many'slots' there are for each. Slots indicate how many processes can potentiallyexecute on a node. For best performance, the number of slots may be chosento be the number of cores on the node or the number of processor sockets. If the hostfile does not provide slots information, Open MPI will attemptto discover the number of cores (or hwthreads, if the use-hwthreads-as-cpusoption is set) and set the number of slots to that value. This default behavioralso occurs when specifying the -host option with a single hostname. Thus,the command

mpirun -H aa ./a.out

launches a number of processes equal tothe number of cores on node aa.

mpirun -hostfile myhostfile ./a.out

willlaunch two processes on each of the three nodes.

mpirun -hostfile myhostfile-host aa ./a.out

will launch two processes, both on node aa.

mpirun -hostfilemyhostfile -host dd ./a.out

will find no hosts to run on and abort with anerror. That is, the specified host dd is not in the specified hostfile.

When running under resource managers (e.g., SLURM, Torque, etc.), Open MPIwill obtain both the hostnames and the number of slots directly from theresource manger.

Specifying Number of Processes

As we have just seen, thenumber of processes to run can be set using the hostfile. Other mechanismsexist.

The number of processes launched can be specified as a multipleof the number of nodes or processor sockets available. For example,

mpirun-H aa,bb -npersocket 2 ./a.out

launches processes 0-3 on node aa and process4-7 on node bb, where aa and bb are both dual-socket nodes. The -npersocketoption also turns on the -bind-to-socket option, which is discussed in a latersection.

mpirun -H aa,bb -npernode 2 ./a.out

launches processes 0-1 on nodeaa and processes 2-3 on node bb.

mpirun -H aa,bb -npernode 1 ./a.out

launchesone process per host node.

mpirun -H aa,bb -pernode ./a.out

is the same as-npernode 1.

Another alternative is to specify the number of processeswith the -np option. Consider now the hostfile

% cat myhostfile
aa slots=4
bb slots=4
cc slots=4

Now,

mpirun -hostfile myhostfile -np 6 ./a.out

will launch processes 0-3on node aa and processes 4-5 on node bb. The remaining slots in the hostfilewill not be used since the -np option indicated that only 6 processes shouldbe launched.

Mapping Processes to Nodes: Using Policies

The examples aboveillustrate the default mapping of process processes to nodes. This mappingcan also be controlled with various mpirun options that describe mappingpolicies.

Consider the same hostfile as above, again with -np 6:

node aa node bb node cc

mpirun 0 1 2 3 4 5

mpirun --map-by node 0 3 1 4 2 5

mpirun -nolocal 0 1 2 3 4 5

The --map-by node option will load balance the processes across the availablenodes, numbering each process in a round-robin fashion.

The -nolocal optionprevents any processes from being mapped onto the local host (in this casenode aa). While mpirun typically consumes few system resources, -nolocalcan be helpful for launching very large jobs where mpirun may actuallyneed to use noticeable amounts of memory and/or processing time.

Just as-np can specify fewer processes than there are slots, it can also oversubscribethe slots. For example, with the same hostfile:

mpirun -hostfile myhostfile-np 14 ./a.out

will launch processes 0-3 on node aa, 4-7 on bb, and 8-11 on cc. It will then add the remaining two processes to whichever nodes it chooses.

One can also specify limits to oversubscription. For example, with thesame hostfile:

mpirun -hostfile myhostfile -np 14 -nooversubscribe ./a.out

will produce an error since -nooversubscribe prevents oversubscription.

Limits to oversubscription can also be specified in the hostfile itself: % cat myhostfile
aa slots=4 max_slots=4
bb max_slots=4
cc slots=4

The max_slots field specifies such a limit. When it does, the slots valuedefaults to the limit. Now:

mpirun -hostfile myhostfile -np 14 ./a.out

causesthe first 12 processes to be launched as before, but the remaining twoprocesses will be forced onto node cc. The other two nodes are protectedby the hostfile against oversubscription by this job.

Using the --nooversubscribeoption can be helpful since Open MPI currently does not get 'max_slots'values from the resource manager.

Of course, -np can also be used with the-H or -host option. For example,

mpirun -H aa,bb -np 8 ./a.out

launches 8 processes. Since only two hosts are specified, after the first two processes aremapped, one to aa and one to bb, the remaining processes oversubscribethe specified hosts.

And here is a MIMD example:

mpirun -H aa -np 1 hostname: -H bb,cc -np 2 uptime

will launch process 0 running hostname on node aaand processes 1 and 2 each running uptime on nodes bb and cc, respectively.

Mapping, Ranking, and Binding: Oh My!

Open MPI employs a three-phase procedurefor assigning process locations and ranks:

mapping

Assigns a default locationto each process

ranking

Assigns an MPI_COMM_WORLD rank value to each process

binding

Constrains each process to run on specific processors

The mappingstep is used to assign a default location to each process based on themapper being employed. Mapping by slot, node, and sequentially results inthe assignment of the processes to the node level. In contrast, mappingby object, allows the mapper to assign the process to an actual objecton each node.

Note: the location assigned to the process is independentof where it will be bound - the assignment is used solely as input to thebinding algorithm.

The mapping of process processes to nodes can be definednot just with general policies but also, if necessary, using arbitrarymappings that cannot be described by a simple policy. One can use the 'sequentialmapper,' which reads the hostfile line by line, assigning processes tonodes in whatever order the hostfile specifies. Use the -mca rmaps seq option. For example, using the same hostfile as before:

mpirun -hostfile myhostfile-mca rmaps seq ./a.out

will launch three processes, one on each of nodesaa, bb, and cc, respectively. The slot counts don’t matter; one processis launched per line on whatever node is listed on the line.

Another wayto specify arbitrary mappings is with a rankfile, which gives you detailedcontrol over process binding as well. Rankfiles are discussed below.

Thesecond phase focuses on the ranking of the process within the job’s MPI_COMM_WORLD. Open MPI separates this from the mapping procedure to allow more flexibilityin the relative placement of MPI processes. This is best illustrated byconsidering the following two cases where we used the —map-by ppr:2:socketoption:

node aa node bb

rank-by core 0 1 ! 2 3 4 5 ! 6 7

rank-by socket 0 2 ! 1 3 4 6 ! 5 7

rank-by socket:span 0 4 ! 1 5 2 6 ! 3 7

Ranking by core and by slot provide the identical result - a simple progressionof MPI_COMM_WORLD ranks across each node. Ranking by socket does a round-robinranking within each node until all processes have been assigned an MCWrank, and then progresses to the next node. Adding the span modifier tothe ranking directive causes the ranking algorithm to treat the entireallocation as a single entity - thus, the MCW ranks are assigned acrossall sockets before circling back around to the beginning.

The binding phaseactually binds each process to a given set of processors. This can improveperformance if the operating system is placing processes suboptimally.For example, it might oversubscribe some multi-core processor sockets, leavingother sockets idle; this can lead processes to contend unnecessarily forcommon resources. Or, it might spread processes out too widely; this canbe suboptimal if application performance is sensitive to interprocess communicationcosts. Binding can also keep the operating system from migrating processesexcessively, regardless of how optimally those processes were placed tobegin with.

No Hw-module Slot 1 Oversubscription Port-group 1

The processors to be used for binding can be identified interms of topological groupings - e.g., binding to an l3cache will bind eachprocess to all processors within the scope of a single L3 cache withintheir assigned location. Thus, if a process is assigned by the mapper toa certain socket, then a —bind-to l3cache directive will cause the processto be bound to the processors that share a single L3 cache within thatsocket.

Alternatively, processes can be assigned to processors based ontheir local rank on a node using the --bind-to cpu-list:ordered option withan associated --cpu-list '0,2,5'. In this example, the first process on a nodewill be bound to cpu 0, the second process on the node will be bound tocpu 2, and the third process on the node will be bound to cpu 5. --bind-towill also accept cpulist:ortered as a synonym to cpu-list:ordered. Notethat an error will result if more processes are assigned to a node thancpus are provided.

To help balance loads, the binding directive uses around-robin method when binding to levels lower than used in the mapper.For example, consider the case where a job is mapped to the socket level,and then bound to core. Each socket will have multiple cores, so if multipleprocesses are mapped to a given socket, the binding algorithm will assigneach process located to a socket to a unique core in a round-robin manner.

Alternatively, processes mapped by l2cache and then bound to socket willsimply be bound to all the processors in the socket where they are located.In this manner, users can exert detailed control over relative MCW ranklocation and binding.

Finally, --report-bindings can be used to report bindings.

As an example, consider a node with two processor sockets, each comprisingfour cores. We run mpirun with -np 4 --report-bindings and the following additionaloptions:

% mpirun ... --map-by core --bind-to core
[...] ... binding child [...,0] to cpus 0001
[...] ... binding child [...,1] to cpus 0002
[...] ... binding child [...,2] to cpus 0004
[...] ... binding child [...,3] to cpus 0008

% mpirun ... --map-by socket --bind-to socket
[...] ... binding child [...,0] to socket 0 cpus 000f
[...] ... binding child [...,1] to socket 1 cpus 00f0
[...] ... binding child [...,2] to socket 0 cpus 000f
[...] ... binding child [...,3] to socket 1 cpus 00f0

% mpirun ... --map-by core:PE=2 --bind-to core
[...] ... binding child [...,0] to cpus 0003
[...] ... binding child [...,1] to cpus 000c
[...] ... binding child [...,2] to cpus 0030
[...] ... binding child [...,3] to cpus 00c0

% mpirun ... --bind-to none

Here, --report-bindings shows the binding of each process as a mask. In thefirst case, the processes bind to successive cores as indicated by themasks 0001, 0002, 0004, and 0008. In the second case, processes bind toall cores on successive sockets as indicated by the masks 000f and 00f0.The processes cycle through the processor sockets in a round-robin fashionas many times as are needed. In the third case, the masks show us that2 cores have been bound per process. In the fourth case, binding is turnedoff and no bindings are reported.

Open MPI’s support for process bindingdepends on the underlying operating system. Therefore, certain processbinding options may not be available on every system.

Process binding canalso be set with MCA parameters. Their usage is less convenient than thatof mpirun options. On the other hand, MCA parameters can be set not onlyon the mpirun command line, but alternatively in a system or user mca-params.conffile or as environment variables, as described in the MCA section below.Some examples include:

mpirun option MCA parameter key value

--map-by core rmaps_base_mapping_policy core
--map-by socket rmaps_base_mapping_policy socket
--rank-by core rmaps_base_ranking_policy core
--bind-to core hwloc_base_binding_policy core
--bind-to socket hwloc_base_binding_policy socket
--bind-to none hwloc_base_binding_policy none

Rankfiles

Rankfiles are text files that specify detailed informationabout how individual processes should be mapped to nodes, and to whichprocessor(s) they should be bound. Each line of a rankfile specifies thelocation of one process (for MPI jobs, the process’ 'rank' refers to itsrank in MPI_COMM_WORLD). The general form of each line in the rankfileis:

rank <N>=<hostname> slot=<slot list>

For example:

$ cat myrankfile
rank 0=aa slot=1:0-2
rank 1=bb slot=0:0,1
rank 2=cc slot=1-2
$ mpirun -H aa,bb,cc,dd -rf myrankfile ./a.out

Means that

Rank 0 runs on node aa, bound to logical socket 1, cores0-2.
Rank 1 runs on node bb, bound to logical socket 0, cores 0 and 1.
Rank 2 runs on node cc, bound to logical cores 1 and 2.

Rankfiles can alternatively be used to specify physical processor locations.In this case, the syntax is somewhat different. Sockets are no longer recognized,and the slot number given must be the number of the physical PU as mostOS’s do not assign a unique physical identifier to each core in the node.Thus, a proper physical rankfile looks something like the following:

$ cat myphysicalrankfile
rank 0=aa slot=1
rank 1=bb slot=8
rank 2=cc slot=6

This means that

Rank 0 will run on node aa, bound to the core thatcontains physical PU 1
Rank 1 will run on node bb, bound to the core that contains physicalPU 8
Rank 2 will run on node cc, bound to the core that contains physicalPU 6

Rankfiles are treated as logical by default, and the MCA parameter rmaps_rank_file_physicalmust be set to 1 to indicate that the rankfile is to be considered as physical.

The hostnames listed above are 'absolute,' meaning that actual resolveablehostnames are specified. However, hostnames can also be specified as 'relative,'meaning that they are specified in relation to an externally-specified listof hostnames (e.g., by mpirun’s --host argument, a hostfile, or a job scheduler).

The 'relative' specification is of the form '+n<X>', where X is an integerspecifying the Xth hostname in the set of all available hostnames, indexedfrom 0. For example:

$ cat myrankfile
rank 0=+n0 slot=1:0-2
rank 1=+n1 slot=0:0,1
rank 2=+n2 slot=1-2
$ mpirun -H aa,bb,cc,dd -rf myrankfile ./a.out

Starting with Open MPI v1.7, all socket/core slot locations are be specifiedas logical indexes (the Open MPI v1.6 series used physical indexes). Youcan use tools such as HWLOC’s 'lstopo' to find the logical indexes of socketand cores.

Application Context or Executable Program?

To distinguish thetwo different forms, mpirun looks on the command line for --app option. Ifit is specified, then the file named on the command line is assumed tobe an application context. If it is not specified, then the file is assumedto be an executable program.

Locating Files

If no relative or absolutepath is specified for a file, Open MPI will first look for files by searchingthe directories specified by the --path option. If there is no --path optionset or if the file is not found at the --path location, then Open MPI willsearch the user’s PATH environment variable as defined on the source node(s).

If a relative directory is specified, it must be relative to the initialworking directory determined by the specific starter used. For example whenusing the rsh or ssh starters, the initial directory is $HOME by default.Other starters may set the initial directory to the current working directoryfrom the invocation of mpirun.

Current Working Directory

The -wdir mpirunoption (and its synonym, -wd) allows the user to change to an arbitrarydirectory before the program is invoked. It can also be used in applicationcontext files to specify working directories on specific nodes and/or forspecific applications.

If the -wdir option appears both in a context fileand on the command line, the context file directory will override the commandline value.

If the -wdir option is specified, Open MPI will attempt to changeto the specified directory on all of the remote nodes. If this fails, mpirunwill abort.

If the -wdir option is not specified, Open MPI will send thedirectory name where mpirun was invoked to each of the remote nodes. Theremote nodes will try to change to that directory. If they are unable (e.g.,if the directory does not exist on that node), then Open MPI will use thedefault directory determined by the starter.

All directory changing occursbefore the user’s program is invoked; it does not wait until MPI_INIT iscalled.

Standard I/O

Open MPI directs UNIX standard input to /dev/nullon all processes except the MPI_COMM_WORLD rank 0 process. The MPI_COMM_WORLDrank 0 process inherits standard input from mpirun. Note: The node thatinvoked mpirun need not be the same as the node where the MPI_COMM_WORLDrank 0 process resides. Open MPI handles the redirection of mpirun’s standardinput to the rank 0 process.

Open MPI directs UNIX standard output and errorfrom remote nodes to the node that invoked mpirun and prints it on thestandard output/error of mpirun. Local processes inherit the standard output/errorof mpirun and transfer to it directly.

Thus it is possible to redirect standardI/O for Open MPI applications by using the typical shell redirection procedureon mpirun.

% mpirun -np 2 my_app < my_input > my_output

Note that in this example only the MPI_COMM_WORLD rank 0 process willreceive the stream from my_input on stdin. The stdin on all the other nodeswill be tied to /dev/null. However, the stdout from all nodes will be collectedinto the my_output file.

Signal Propagation

When orterun receives a SIGTERMand SIGINT, it will attempt to kill the entire job by sending all processesin the job a SIGTERM, waiting a small number of seconds, then sending allprocesses in the job a SIGKILL.

SIGUSR1 and SIGUSR2 signals received byorterun are propagated to all processes in the job.

A SIGTSTOP signal tompirun will cause a SIGSTOP signal to be sent to all of the programs startedby mpirun and likewise a SIGCONT signal to mpirun will cause a SIGCONTsent.

Other signals are not currently propagated by orterun.

Process Termination/ Signal Handling

During the run of an MPI application, if any processdies abnormally (either exiting before invoking MPI_FINALIZE, or dyingas the result of a signal), mpirun will print out an error message andkill the rest of the MPI application.

User signal handlers should probablyavoid trying to cleanup MPI state (Open MPI is currently not async-signal-safe;see MPI_Init_thread(3) for details about MPI_THREAD_MULTIPLE and threadsafety). For example, if a segmentation fault occurs in MPI_SEND (perhapsbecause a bad buffer was passed in) and a user signal handler is invoked,if this user handler attempts to invoke MPI_FINALIZE, Bad Things couldhappen since Open MPI was already 'in' MPI when the error occurred. Sincempirun will notice that the process died due to a signal, it is probablynot necessary (and safest) for the user to only clean up non-MPI state.

Process Environment

Processes in the MPI application inherit their environmentfrom the Open RTE daemon upon the node on which they are running. The environmentis typically inherited from the user’s shell. On remote nodes, the exactenvironment is determined by the boot MCA module used. The rsh launch module,for example, uses either rsh/ssh to launch the Open RTE daemon on remotenodes, and typically executes one or more of the user’s shell-setup filesbefore launching the Open RTE daemon. When running dynamically linked applicationswhich require the LD_LIBRARY_PATH environment variable to be set, caremust be taken to ensure that it is correctly set when booting Open MPI.

See the 'Remote Execution' section for more details.

Remote Execution

Open MPI requires that the PATH environment variable be set to find executableson remote nodes (this is typically only necessary in rsh- or ssh-based environments-- batch/scheduled environments typically copy the current environment tothe execution of remote jobs, so if the current environment has PATH and/orLD_LIBRARY_PATH set properly, the remote nodes will also have it set properly). If Open MPI was compiled with shared library support, it may also be necessaryto have the LD_LIBRARY_PATH environment variable set on remote nodes aswell (especially to find the shared libraries required to run user MPIapplications).

However, it is not always desirable or possible to edit shellstartup files to set PATH and/or LD_LIBRARY_PATH. The --prefix option isprovided for some simple configurations where this is not possible.

The--prefix option takes a single argument: the base directory on the remotenode where Open MPI is installed. Open MPI will use this directory to setthe remote PATH and LD_LIBRARY_PATH before executing any Open MPI or userapplications. This allows running Open MPI jobs without having pre-configuredthe PATH and LD_LIBRARY_PATH on the remote nodes.

Open MPI adds the basenameof the current node’s 'bindir' (the directory where Open MPI’s executablesare installed) to the prefix and uses that to set the PATH on the remotenode. Similarly, Open MPI adds the basename of the current node’s 'libdir'(the directory where Open MPI’s libraries are installed) to the prefix anduses that to set the LD_LIBRARY_PATH on the remote node. For example:

Localbindir:

/local/node/directory/bin

Local libdir:

/local/node/directory/lib64

If the following command line is used:

% mpirun --prefix /remote/node/directory

Open MPI will add '/remote/node/directory/bin' to the PATH and '/remote/node/directory/lib64'to the LD_LIBRARY_PATH on the remote node before attempting to executeanything.

The --prefix option is not sufficient if the installation pathson the remote node are different than the local node (e.g., if '/lib' isused on the local node, but '/lib64' is used on the remote node), or ifthe installation paths are something other than a subdirectory under acommon prefix.

Note that executing mpirun via an absolute pathname is equivalentto specifying --prefix without the last subdirectory in the absolute pathnameto mpirun. For example:

% /usr/local/bin/mpirun ...

is equivalent to

% mpirun --prefix /usr/local

Exported Environment Variables

All environment variables that are namedin the form OMPI_* will automatically be exported to new processes on thelocal and remote nodes. Environmental parameters can also be set/forwardedto the new processes using the MCA parameter mca_base_env_list. The -x optionto mpirun has been deprecated, but the syntax of the MCA param followsthat prior example. While the syntax of the -x option and MCA param allowsthe definition of new variables, note that the parser for these optionsare currently not very sophisticated - it does not even understand quotedvalues. Users are advised to set variables in the environment and use theoption to export them; not to define them.

Setting MCA Parameters

The-mca switch allows the passing of parameters to various MCA (Modular ComponentArchitecture) modules. MCA modules have direct impact on MPI programs becausethey allow tunable parameters to be set at run time (such as which BTLcommunication device driver to use, what parameters to pass to that BTL,etc.).

The -mca switch takes two arguments: <key> and <value>. The <key> argumentgenerally specifies which MCA module will receive the value. For example,the <key> 'btl' is used to select which BTL to be used for transporting MPImessages. The <value> argument is the value that is passed. For example:

mpirun -mca btl tcp,self -np 1 foo

Tells Open MPI to use the 'tcp' and 'self'BTLs, and to run a single copy of 'foo' an allocated node.

mpirun -mca btlself -np 1 foo

Tells Open MPI to use the 'self' BTL, and to run a singlecopy of 'foo' an allocated node.

The -mca switch can be used multiple timesto specify different <key> and/or <value> arguments. If the same <key> is specifiedmore than once, the <value>s are concatenated with a comma (',') separatingthem.

Note that the -mca switch is simply a shortcut for setting environmentvariables. The same effect may be accomplished by setting correspondingenvironment variables before running mpirun. The form of the environmentvariables that Open MPI sets is:

OMPI_MCA_<key>=<value>

Thus, the -mca switch overrides any previously set environment variables. The -mca settings similarly override MCA parameters set in the $OPAL_PREFIX/etc/openmpi-mca-params.confor $HOME/.openmpi/mca-params.conf file.

Unknown <key> arguments are still setas environment variable -- they are not checked (by mpirun) for correctness.Illegal or incorrect <value> arguments may or may not be reported -- it dependson the specific MCA module.

To find the available component types underthe MCA architecture, or to find the available parameters for a specificcomponent, use the ompi_info command. See the ompi_info(1) man page fordetailed information on the command.

Setting MCA parameters and environmentvariables from file.

The -tune command line option and its synonym -mca mca_base_envar_file_prefix

No Hw-module Switch 1 Slot 1 Oversubscription Port-group 1

allows a user to set mca parameters and environment variables with thesyntax described below. This option requires a single file or list of filesseparated by ',' to follow.

A valid line in the file may contain zero ormany '-x', '-mca', or “--mca” arguments. The following patterns are supported:-mca var val -mca var 'val' -x var=val -x var. If any argument is duplicatedin the file, the last value read will be used.

MCA parameters and environmentspecified on the command line have higher precedence than variables specifiedin the file.

Running as root

The Open MPI team strongly advises againstexecuting mpirun as the root user. MPI applications should be run as regular(non-root) users.

Reflecting this advice, mpirun will refuse to run as rootby default. To override this default, you can add the --allow-run-as-root optionto the mpirun command line, or you can set the environmental parametersOMPI_ALLOW_RUN_AS_ROOT=1 and OMPI_ALLOW_RUN_AS_ROOT_CONFIRM=1. Note thatit takes setting two environment variables to effect the same behavioras --allow-run-as-root in order to stress the Open MPI team’s strong advice againstrunning as the root user. After extended discussions with communities whouse containers (where running as the root user is the default), there wasa persistent desire to be able to enable root execution of mpirun via anenvironmental control (vs. the existing --allow-run-as-root command line parameter). The compromise of using two environment variables was reached: it allowsroot execution via an environmental control, but it conveys the Open MPIteam’s strong recomendation against this behavior.

Exit status

There isno standard definition for what mpirun should return as an exit status.After considerable discussion, we settled on the following method for assigningthe mpirun exit status (note: in the following description, the 'primary'job is the initial application started by mpirun - all jobs that are spawnedby that job are designated 'secondary' jobs):

[bu]

if all processes inthe primary job normally terminate with exit status 0, we return 0

[bu]

ifone or more processes in the primary job normally terminate with non-zeroexit status, we return the exit status of the process with the lowest MPI_COMM_WORLDrank to have a non-zero status

[bu]

if all processes in the primary job normallyterminate with exit status 0, and one or more processes in a secondaryjob normally terminate with non-zero exit status, we (a) return the exitstatus of the process with the lowest MPI_COMM_WORLD rank in the lowestjobid to have a non-zero status, and (b) output a message summarizing theexit status of the primary and all secondary jobs.

[bu]

if the cmd line option--report-child-jobs-separately is set, we will return -only- the exit status ofthe primary job. Any non-zero exit status in secondary jobs will be reportedsolely in a summary print statement.

By default, OMPI records and notesthat MPI processes exited with non-zero termination status. This is generallynot considered an 'abnormal termination' - i.e., OMPI will not abort an MPIjob if one or more processes return a non-zero status. Instead, the defaultbehavior simply reports the number of processes terminating with non-zerostatus upon completion of the job.

However, in some cases it can be desirableto have the job abort when any process terminates with non-zero status. Forexample, a non-MPI job might detect a bad result from a calculation andwant to abort, but doesn’t want to generate a core file. Or an MPI job mightcontinue past a call to MPI_Finalize, but indicate that all processes shouldabort due to some post-MPI result.

It is not anticipated that this situationwill occur frequently. However, in the interest of serving the broader community,OMPI now has a means for allowing users to direct that jobs be abortedupon any process exiting with non-zero status. Setting the MCA parameter'orte_abort_on_non_zero_status' to 1 will cause OMPI to abort all processesonce any process exits with non-zero status.

Terminations caused in this manner will be reported on the console asan 'abnormal termination', with the first process to so exit identifiedalong with its exit status.

Examples

Be sure also to see the examplesthroughout the sections above.

mpirun -np 4 -mca btl ib,tcp,self prog1

Run4 copies of prog1 using the 'ib', 'tcp', and 'self' BTL’s for the transportof MPI messages.

mpirun -np 4 -mca btl tcp,sm,self

--mca btl_tcp_if_include eth0 prog1
Run 4 copies of prog1 using the 'tcp', 'sm' and 'self' BTLs for the transportof MPI messages, with TCP using only the eth0 interface to communicate. Note that other BTLs have similar if_include MCA parameters.

Return Value

mpirun returns 0 if all processes started by mpirun exitafter calling MPI_FINALIZE. A non-zero value is returned if an internalerror occurred in mpirun, or one or more processes exited before callingMPI_FINALIZE. If an internal error occurred in mpirun, the correspondingerror code is returned. In the event that one or more processes exit beforecalling MPI_FINALIZE, the return value of the MPI_COMM_WORLD rank of theprocess that mpirun first notices died before calling MPI_FINALIZE willbe returned. Note that, in general, this will be the first process thatdied but is not guaranteed to be so.

If the --timeout command line optionis used and the timeout expires before the job completes (thereby forcingmpirun to kill the job) mpirun will return an exit status equivalent tothe value of ETIMEDOUT (which is typically 110 on Linux and OS X systems).

See Also

MPI_Init_thread(3)