NOTE: launch performance will be absolutely awful if you do this with BTLs that aren't configured to modex_recv on first message!
Even with "modex on demand", we still have to do a barrier in place of the modex - we simply don't move any data around, which does reduce the time impact. The barrier is required to ensure that the other proc has in fact registered all its BTL info and therefore is prepared to hand over a complete data package. Otherwise, you may not get the info you need. In addition, the shared memory BTL can fail to properly rendezvous as it expects the barrier to be in place.
This behavior will *only* take effect under the following conditions:
1. launched via mpirun
2. #procs is greater than ompi_hostname_cutoff, which defaults to UINT32_MAX
3. mca param rte_orte_direct_modex is set to 1. At the moment, we are having problems getting this param to register properly, so only the first two conditions are in effect. Still, the bottom line is you have to *want* this behavior to get it.
The planned next evolution of this will be to make the direct modex be non-blocking - this will require two fixes:
1. if the remote proc doesn't have the required info, then let it delay its response until it does. This means we need a way for the MPI layer to tell the RTE "I am done entering modex data".
2. adjust the SM rendezvous logic to loop until the required file has been created
Creating a placeholder to bring this over to 1.7.5 when ready.
cmr=v1.7.5:reviewer=hjelmn:subject=Enable direct modex at scale
This commit was SVN r30259.
So shift the cutoff param to the MPI layer, and have it solely determine whether or not we call modex_recv on the hostname. If comm_world is of size greater than the cutoff, then we don't automatically retrieve the hostname when we build the ompi_proc_t for a process - instead, we fill the hostname entry on first call to modex_recv for that process.
The param is now "ompi_hostname_cutoff=N", where N=number of procs for cutoff.
Refs trac:3729
This commit was SVN r29056.
The following Trac tickets were found above:
Ticket 3729 --> https://svn.open-mpi.org/trac/ompi/ticket/3729
* add a new MCA param orte_hostname_cutoff to specify the number of nodes at which we stop including hostnames. This defaults to INT_MAX => always include hostnames. If a value is given, then we will include hostnames for any allocation smaller than the given limit.
* remove ompi_proc_get_hostname. Replace all occurrences with a direct link to ompi_proc_t's proc_hostname, protected by appropriate "if NULL"
* modify the OMPI-ORTE integration component so that any call to modex_recv automatically loads the ompi_proc_t->proc_hostname field as well as returning the requested info. Thus, any process whose modex info you retrieve will automatically receive the hostname. Note that on-demand retrieval is still enabled - i.e., if we are running under direct launch with PMI, the hostname will be fetched upon first call to modex_recv, and then the ompi_proc_t->proc_hostname field will be loaded
* removed a stale MCA param "mpi_keep_peer_hostnames" that was no longer used anywhere in the code base
* added an envar lookup in ess/pmi for the number of nodes in the allocation. Sadly, PMI itself doesn't provide that info, so we have to get it a different way. Currently, we support PBS-based systems and SLURM - for any other, rank0 will emit a warning and we assume max number of daemons so we will always retain hostnames
This commit was SVN r29052.
Features:
- Support for an override parameter file (openmpi-mca-param-override.conf).
Variable values in this file can not be overridden by any file or environment
value.
- Support for boolean, unsigned, and unsigned long long variables.
- Support for true/false values.
- Support for enumerations on integer variables.
- Support for MPIT scope, verbosity, and binding.
- Support for command line source.
- Support for setting variable source via the environment using
OMPI_MCA_SOURCE_<var name>=source (either command or file:filename)
- Cleaner API.
- Support for variable groups (equivalent to MPIT categories).
Notes:
- Variables must be created with a backing store (char **, int *, or bool *)
that must live at least as long as the variable.
- Creating a variable with the MCA_BASE_VAR_FLAG_SETTABLE enables the use of
mca_base_var_set_value() to change the value.
- String values are duplicated when the variable is registered. It is up to
the caller to free the original value if necessary. The new value will be
freed by the mca_base_var system and must not be freed by the user.
- Variables with constant scope may not be settable.
- Variable groups (and all associated variables) are deregistered when the
component is closed or the component repository item is freed. This
prevents a segmentation fault from accessing a variable after its component
is unloaded.
- After some discussion we decided we should remove the automatic registration
of component priority variables. Few component actually made use of this
feature.
- The enumerator interface was updated to be general enough to handle
future uses of the interface.
- The code to generate ompi_info output has been moved into the MCA variable
system. See mca_base_var_dump().
opal: update core and components to mca_base_var system
orte: update core and components to mca_base_var system
ompi: update core and components to mca_base_var system
This commit also modifies the rmaps framework. The following variables were
moved from ppr and lama: rmaps_base_pernode, rmaps_base_n_pernode,
rmaps_base_n_persocket. Both lama and ppr create synonyms for these variables.
This commit was SVN r28236.
ompi_show_help, because opal_show_help is replaced with an
aggregating version when using ORTE, so there's no reason to
directly call orte_show_help.
This commit was SVN r28051.
1. replacing mpi_paffinity_alone with opal_paffinity_alone - for back-compatibility, I have aliased mpi_paffinity_alone to the new param name. This caus
es a mild abstraction break in the opal/mca/paffinity framework - per the devel discussion...live with it. :-) I also moved the ompi_xxx global variable
that tracked maffinity setup so it could be properly closed in MPI_Finalize to the opal/mca/maffinity framework to avoid an abstraction break.
2. Added code to the odls/default module to perform paffinity binding and maffinity init between process fork and exec. This has been tested on IU's odi
n cluster and works for both MPI and non-MPI apps.
3. Revise MPI_Init to detect if affinity has already been set, and to attempt to set it if not already done. I have *not* tested this as I haven't yet f
igured out a way to do so - I couldn't get slurm to perform cpu bindings, even though it supposedly does do so.
This has only been lightly tested and would definitely benefit from a wider range of evaluation...
This commit was SVN r21209.
OMPI_* to OPAL_*. This allows opal layer to be used more independent
from the whole of ompi.
NOTE: 9 "svn mv" operations immediately follow this commit.
This commit was SVN r21180.
- Delete unnecessary header files using
contrib/check_unnecessary_headers.sh after applying
patches, that include headers, being "lost" due to
inclusion in one of the now deleted headers...
In total 817 files are touched.
In ompi/mpi/c/ header files are moved up into the actual c-file,
where necessary (these are the only additional #include),
otherwise it is only deletions of #include (apart from the above
additions required due to notifier...)
- To get different MCAs (OpenIB, TM, ALPS), an earlier version was
successfully compiled (yesterday) on:
Linux locally using intel-11, gcc-4.3.2 and gcc-SVN + warnings enabled
Smoky cluster (x86-64 running Linux) using PGI-8.0.2 + warnings enabled
Lens cluster (x86-64 running Linux) using Pathscale-3.2 + warnings enabled
This commit was SVN r21096.
several header files (previously included by header-files)
now have to be moved "upward".
This is mainly system headers such as string.h, stdio.h and for
networking, but also some orte headers.
This commit was SVN r21095.
only show up if a notifier component is selected, of course). These
can be disabled by setting the MCA parameter mpi_notify_init_finalize
to 0.
These messages are both intended as "hey, does the community like
this?" and as a way to get some real-world testing of the notify
system. The default is currently to send these messages if a notify
component is selected; we can change the default later if desired.
This commit was SVN r21078.
In case we use memcmp, strlen, strup and friends include <string.h>
Also several constants.h are not included directly
- Let's have mca_topo_base_cart_create return ompi-errors in
ompi/mca/topo/base/topo_base_cart_create.c
This commit was SVN r20773.
Often, orte/util/show_help.h is included, although no functionality
is required -- instead, most often opal_output.h, or
orte/mca/rml/rml_types.h
Please see orte_show_help_replacement.sh commited next.
- Local compilation (Linux/x86_64) w/ -Wimplicit-function-declaration
actually showed two *missing* #include "orte/util/show_help.h"
in orte/mca/odls/base/odls_base_default_fns.c and
in orte/tools/orte-top/orte-top.c
Manually added these.
Let's have MTT the last word.
This commit was SVN r20557.
is >= 1. The default value of the MCA param is now -1, which means
"let someone else turn it on if they want to." So we should default
to ''off'' (false), and let the openib BTL (etc.) turn it on if it
can/wants to.
Failure to do this will default _pipeline to true because
-1(int)==true(bool). This causes a problem if the user tries to set
mpi_leave_pinned_pipeline to 1: they'll get a warning that you can't
set both _pinned and _pinned_pipeline to 1. This happens because
_pinned will get the bool-ified value of of the MCA parameter (-1),
and then the user sets the value of _pinned_pipeline to 1/true.
Hence, both of them are set to true. Bzzt!
This commit was SVN r19942.
1. register "mpi_preconnect_all" as a deprecated synonym for "mpi_preconnect_mpi"
2. remove "mpi_preconnect_oob" and "mpi_preconnect_oob_simultaneous" as these are no longer valid.
3. remove the routed framework's "warmup_routes" API. With the removal of the direct routed component, this function at best only wasted communications. The daemon routes are completely "warmed up" during launch, so having MPI procs order the sending of additional messages is simply wasteful.
4. remove the call to orte_routed.warmup_routes from MPI_Init. This was the only place it was used anyway.
The FAQs will be updated to reflect this changed situation, and a CMR filed to move this to the 1.3 branch.
This commit was SVN r19933.
figure it out at runtime (really meaning: we'll still default to "0"
unless something explicitly overrides to 1, such as the openib BTL).
This way, ompi_info doesn't confusingly report mpi_leave_pinned==0 for
mpi_leave_pinned, but we end up running with mpi_leave_pinned==1.
Fixes trac:1502.
This commit was SVN r19571.
The following Trac tickets were found above:
Ticket 1502 --> https://svn.open-mpi.org/trac/ompi/ticket/1502
* use "warn_on_fork" instead of "do_not_warn_on_fork" -- i.e.,
use positive logic instead of negative logic
* ensure that pthread_atfork() is only called once
* amended the error message to include the hostname, PID, and
MPI_COMM_WORLD rank of the offender
* ensure that the warn_fork_cb() function is only defined if
HAVE_PTHREAD_H so that we don't get a compiler warning if it isn't
used
This commit was SVN r19204.
The following SVN revision numbers were found above:
r19196 --> open-mpi/ompi@277e4ac292
This is setup so that it only is issued once (as opposed to every time they do it), and goes through orte_show_help so the user doesn't get hammered by #procs copies of the warning. In addition, there is a new MCA param (can't have too many!) to shut the warning off altogether.
This closes ticket #1244
This commit was SVN r19196.
put the name of the file that set them if they were set by file. This is of great assistance to support personnel trying to understand why a user is having pro
blems.
Coordinated with Jeff.
This commit was SVN r19111.
Short version: remove opal_paffinity_alone and restore
mpi_paffinity_alone. ORTE makes various information available for the
MPI layer to decide what it wants to do in terms of processor
affinity.
Details:
* remove opal_paffinity_alone MCA param; restore mpi_paffinity_alone
MCA param
* move opal_paffinity_slot_list param registration to paffinity base
* ompi_mpi_init() calls opal_paffinity_base_slot_list_set(); if that
succeeds use that. If no slot list was set, see if
mpi_paffinity_alone was set. If so, bind this process to its Node
Local Rank (NLR). The NLR is the ORTE-maintained slot ID; if you
COMM_SPAWN to a host in this ORTE universe that already has procs
on it, the NLR for the new job will start at N (not 0). So this is
slightly better than mpi_paffinity_alone in the v1.2 series.
* If a slot list is specified *and* mpi_paffinity_alone is set, we
display an error and abort.
* Remove calls from rmaps/rank_file component to register and lookup
opal_paffinity mca params.
* Remove code in orte/odls that set affinities - instead, have them
just pass a slot_list if it exists.
* Cleanup the orte/odls code that determined
oversubscribed/want_processor as these were just opposites of each
other.
This commit was SVN r18874.
The following Trac tickets were found above:
Ticket 1383 --> https://svn.open-mpi.org/trac/ompi/ticket/1383
Lenny and I went back and forth on whether we should simply register
another "mpi_paffinity_alone" MCA param and then try to figure out
which one was set in ompi_mpi_init, but there was difficulty in
figuring out what to do. So it seemed like the Right Thing to do was
to implement what was committed in r18770; then we could tell where
MCA parameters were set from and you could do Better Things (this is
also useful in the openib BTL, where parameters can be set either via
MCA parameter or via an INI file).
But after that was done, it seemed only a few steps further to
actually implement two new features in the MCA params area:
* Synonyms (where one MCA param name is a synonym for another)
* Allow MCA params and/or their synonyms to be marked as "deprecated"
(printing out warnings if they are used)
These features have actually long been discussed/desired, and I had
some time in airports and airplanes recently where I could work in
this stuff on a standalone laptop. So I did it. :-)
This commit introduces these two new features, and then uses them to
register mpi_paffinity_alone as a non-deprecated synonym for
opal_paffinity_alone. A few other random points in this commit:
* Add a few error checks for conditions that were not checked before
* Correct some comments in mca_base_params.h
* Add a few comments in strategic places
* ompi_info now prints additional information:
* for any MCA parameter that has synonyms, it lists all the
synonyms
* synonyms are also output as 1st-class MCA params, but with an
additional attribute indicating that they have a "parent"
* all MCA param name (both "real" or "synonym") will output an
attribute indicating whether it is deprecated or not. A synonym
is deprecated if it iself is marked as deprecated (via the
mca_base_param_regist_syn() or mca_base_param_register_syn_name()
functions) or if its "parent" MCA parameter is deprecated
This commit was SVN r18859.
The following SVN revision numbers were found above:
r18770 --> open-mpi/ompi@8efe67e08c
The following Trac tickets were found above:
Ticket 1383 --> https://svn.open-mpi.org/trac/ompi/ticket/1383
After much work by Jeff and myself, and quite a lot of discussion, it has become clear that we simply cannot resolve the infinite loops caused by RML-involved subsystems calling orte_output. The original rationale for the change to orte_output has also been reduced by shifting the output of XML-formatted vs human readable messages to an alternative approach.
I have globally replaced the orte_output/ORTE_OUTPUT calls in the code base, as well as the corresponding .h file name. I have test compiled and run this on the various environments within my reach, so hopefully this will prove minimally disruptive.
This commit was SVN r18619.
such, the commit message back to the master SVN repository is fairly
long.
= ORTE Job-Level Output Messages =
Add two new interfaces that should be used for all new code throughout
the ORTE and OMPI layers (we already make the search-and-replace on
the existing ORTE / OMPI layers):
* orte_output(): (and corresponding friends ORTE_OUTPUT,
orte_output_verbose, etc.) This function sends the output directly
to the HNP for processing as part of a job-specific output
channel. It supports all the same outputs as opal_output()
(syslog, file, stdout, stderr), but for stdout/stderr, the output
is sent to the HNP for processing and output. More on this below.
* orte_show_help(): This function is a drop-in-replacement for
opal_show_help(), with two differences in functionality:
1. the rendered text help message output is sent to the HNP for
display (rather than outputting directly into the process' stderr
stream)
1. the HNP detects duplicate help messages and does not display them
(so that you don't see the same error message N times, once from
each of your N MPI processes); instead, it counts "new" instances
of the help message and displays a message every ~5 seconds when
there are new ones ("I got X new copies of the help message...")
opal_show_help and opal_output still exist, but they only output in
the current process. The intent for the new orte_* functions is that
they can apply job-level intelligence to the output. As such, we
recommend that all new ORTE and OMPI code use the new orte_*
functions, not thei opal_* functions.
=== New code ===
For ORTE and OMPI programmers, here's what you need to do differently
in new code:
* Do not include opal/util/show_help.h or opal/util/output.h.
Instead, include orte/util/output.h (this one header file has
declarations for both the orte_output() series of functions and
orte_show_help()).
* Effectively s/opal_output/orte_output/gi throughout your code.
Note that orte_output_open() takes a slightly different argument
list (as a way to pass data to the filtering stream -- see below),
so you if explicitly call opal_output_open(), you'll need to
slightly adapt to the new signature of orte_output_open().
* Literally s/opal_show_help/orte_show_help/. The function signature
is identical.
=== Notes ===
* orte_output'ing to stream 0 will do similar to what
opal_output'ing did, so leaving a hard-coded "0" as the first
argument is safe.
* For systems that do not use ORTE's RML or the HNP, the effect of
orte_output_* and orte_show_help will be identical to their opal
counterparts (the additional information passed to
orte_output_open() will be lost!). Indeed, the orte_* functions
simply become trivial wrappers to their opal_* counterparts. Note
that we have not tested this; the code is simple but it is quite
possible that we mucked something up.
= Filter Framework =
Messages sent view the new orte_* functions described above and
messages output via the IOF on the HNP will now optionally be passed
through a new "filter" framework before being output to
stdout/stderr. The "filter" OPAL MCA framework is intended to allow
preprocessing to messages before they are sent to their final
destinations. The first component that was written in the filter
framework was to create an XML stream, segregating all the messages
into different XML tags, etc. This will allow 3rd party tools to read
the stdout/stderr from the HNP and be able to know exactly what each
text message is (e.g., a help message, another OMPI infrastructure
message, stdout from the user process, stderr from the user process,
etc.).
Filtering is not active by default. Filter components must be
specifically requested, such as:
{{{
$ mpirun --mca filter xml ...
}}}
There can only be one filter component active.
= New MCA Parameters =
The new functionality described above introduces two new MCA
parameters:
* '''orte_base_help_aggregate''': Defaults to 1 (true), meaning that
help messages will be aggregated, as described above. If set to 0,
all help messages will be displayed, even if they are duplicates
(i.e., the original behavior).
* '''orte_base_show_output_recursions''': An MCA parameter to help
debug one of the known issues, described below. It is likely that
this MCA parameter will disappear before v1.3 final.
= Known Issues =
* The XML filter component is not complete. The current output from
this component is preliminary and not real XML. A bit more work
needs to be done to configure.m4 search for an appropriate XML
library/link it in/use it at run time.
* There are possible recursion loops in the orte_output() and
orte_show_help() functions -- e.g., if RML send calls orte_output()
or orte_show_help(). We have some ideas how to fix these, but
figured that it was ok to commit before feature freeze with known
issues. The code currently contains sub-optimal workarounds so
that this will not be a problem, but it would be good to actually
solve the problem rather than have hackish workarounds before v1.3 final.
This commit was SVN r18434.
Update the rsh tree spawn capability so we spawn the next wave of daemons before launching our own local procs.
Add an ability to encode nodenames for large clusters with contiguous node name numbering schemes - this allows communication of all node names in a few bytes instead of tens-of-bytes/node.
This commit was SVN r18338.
1. applied prefix rule to functions and variables of RMAPS rank_file component
2. cleaned ompi_mpi_init.c from paffinity code
3. paffinity code moved to new opal/mca/paffinity/base/paffinity_base_service.c file
4. added opal_paffinity_slot_list mca parameter
This commit was SVN r18019.
The commit has been tested for C/R and Cray operations, and on Odin (SLURM, rsh) and RoadRunner (TM). I tried to update all environments, but obviously could not test them. I know that Windows needs some work, and have highlighted what is know to be needed in the odls process component.
This represents a lot of work by Brian, Tim P, Josh, and myself, with much advice from Jeff and others. For posterity, I have appended a copy of the email describing the work that was done:
As we have repeatedly noted, the modex operation in MPI_Init is the single greatest consumer of time during startup. To-date, we have executed that operation as an ORTE stage gate that held the process until a startup message containing all required modex (and OOB contact info - see #3 below) info could be sent to it. Each process would send its data to the HNP's registry, which assembled and sent the message when all processes had reported in.
In addition, ORTE had taken responsibility for monitoring process status as it progressed through a series of "stage gates". The process reported its status at each gate, and ORTE would then send a "release" message once all procs had reported in.
The incoming changes revamp these procedures in three ways:
1. eliminating the ORTE stage gate system and cleanly delineating responsibility between the OMPI and ORTE layers for MPI init/finalize. The modex stage gate (STG1) has been replaced by a collective operation in the modex itself that performs an allgather on the required modex info. The allgather is implemented using the orte_grpcomm framework since the BTL's are not active at that point. At the moment, the grpcomm framework only has a "basic" component analogous to OMPI's "basic" coll framework - I would recommend that the MPI team create additional, more advanced components to improve performance of this step.
The other stage gates have been replaced by orte_grpcomm barrier functions. We tried to use MPI barriers instead (since the BTL's are active at that point), but - as we discussed on the telecon - these are not currently true barriers so the job would hang when we fell through while messages were still in process. Note that the grpcomm barrier doesn't actually resolve that problem, but Brian has pointed out that we are unlikely to ever see it violated. Again, you might want to spend a little time on an advanced barrier algorithm as the one in "basic" is very simplistic.
Summarizing this change: ORTE no longer tracks process state nor has direct responsibility for synchronizing jobs. This is now done via collective operations within the MPI layer, albeit using ORTE collective communication services. I -strongly- urge the MPI team to implement advanced collective algorithms to improve the performance of this critical procedure.
2. reducing the volume of data exchanged during modex. Data in the modex consisted of the process name, the name of the node where that process is located (expressed as a string), plus a string representation of all contact info. The nodename was required in order for the modex to determine if the process was local or not - in addition, some people like to have it to print pretty error messages when a connection failed.
The size of this data has been reduced in three ways:
(a) reducing the size of the process name itself. The process name consisted of two 32-bit fields for the jobid and vpid. This is far larger than any current system, or system likely to exist in the near future, can support. Accordingly, the default size of these fields has been reduced to 16-bits, which means you can have 32k procs in each of 32k jobs. Since the daemons must have a vpid, and we require one daemon/node, this also restricts the default configuration to 32k nodes.
To support any future "mega-clusters", a configuration option --enable-jumbo-apps has been added. This option increases the jobid and vpid field sizes to 32-bits. Someday, if necessary, someone can add yet another option to increase them to 64-bits, I suppose.
(b) replacing the string nodename with an integer nodeid. Since we have one daemon/node, the nodeid corresponds to the local daemon's vpid. This replaces an often lengthy string with only 2 (or at most 4) bytes, a substantial reduction.
(c) when the mca param requesting that nodenames be sent to support pretty error messages, a second mca param is now used to request FQDN - otherwise, the domain name is stripped (by default) from the message to save space. If someone wants to combine those into a single param somehow (perhaps with an argument?), they are welcome to do so - I didn't want to alter what people are already using.
While these may seem like small savings, they actually amount to a significant impact when aggregated across the entire modex operation. Since every proc must receive the modex data regardless of the collective used to send it, just reducing the size of the process name removes nearly 400MBytes of communication from a 32k proc job (admittedly, much of this comm may occur in parallel). So it does add up pretty quickly.
3. routing RML messages to reduce connections. The default messaging system remains point-to-point - i.e., each proc opens a socket to every proc it communicates with and sends its messages directly. A new option uses the orteds as routers - i.e., each proc only opens a single socket to its local orted. All messages are sent from the proc to the orted, which forwards the message to the orted on the node where the intended recipient proc is located - that orted then forwards the message to its local proc (the recipient). This greatly reduces the connection storm we have encountered during startup.
It also has the benefit of removing the sharing of every proc's OOB contact with every other proc. The orted routing tables are populated during launch since every orted gets a map of where every proc is being placed. Each proc, therefore, only needs to know the contact info for its local daemon, which is passed in via the environment when the proc is fork/exec'd by the daemon. This alone removes ~50 bytes/process of communication that was in the current STG1 startup message - so for our 32k proc job, this saves us roughly 32k*50 = 1.6MBytes sent to 32k procs = 51GBytes of messaging.
Note that you can use the new routing method by specifying -mca routed tree - if you so desire. This mode will become the default at some point in the future.
There are a few minor additional changes in the commit that I'll just note in passing:
* propagation of command line mca params to the orteds - fixes ticket #1073. See note there for details.
* requiring of "finalize" prior to "exit" for MPI procs - fixes ticket #1144. See note there for details.
* cleanup of some stale header files
This commit was SVN r16364.
mpi_show_mpi_alloc_mem_leaks
When activated, MPI_FINALIZE displays a list of memory allocations
from MPI_ALLOC_MEM that were not freed by MPI_FREE_MEM (in each MPI
process).
* If set to a positive integer, display only that many leaks.
* If set to a negative integer, display all leaks.
* If set to 0, do not show any leaks.
This commit was SVN r15736.