54b2cf747e
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 The modex stage gate (STG1) has been replaced by a collective operation in the modex itself that performs an allgather on the required modex info.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 The other stage gates have been replaced by orte_grpcomm barrier functions.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.
98 строки
3.4 KiB
Plaintext
98 строки
3.4 KiB
Plaintext
# -*- text -*-
|
|
#
|
|
# Copyright (c) 2004-2005 The Trustees of Indiana University and Indiana
|
|
# University Research and Technology
|
|
# Corporation. All rights reserved.
|
|
# Copyright (c) 2004-2005 The University of Tennessee and The University
|
|
# of Tennessee Research Foundation. All rights
|
|
# reserved.
|
|
# Copyright (c) 2004-2005 High Performance Computing Center Stuttgart,
|
|
# University of Stuttgart. All rights reserved.
|
|
# Copyright (c) 2004-2005 The Regents of the University of California.
|
|
# All rights reserved.
|
|
# $COPYRIGHT$
|
|
#
|
|
# Additional copyrights may follow
|
|
#
|
|
# $HEADER$
|
|
#
|
|
# This is the US/English general help file for Open RTE's orterun.
|
|
#
|
|
[bproc-vexecmove-launch]
|
|
The bproc PLS component was not able to launch %s on node %d and therefore
|
|
cannot continue. Errno was set to %d.
|
|
[bproc-vexecmove-fork]
|
|
The bproc PLS component was not able to fork and therefore cannot continue.
|
|
Errno was set to %d.
|
|
[no-orted]
|
|
The bproc PLS component was not able to find the executable "%s" in the
|
|
current directory, your PATH, or in the directory where Open MPI was
|
|
initially installed, and therefore cannot continue.
|
|
|
|
For reference, we looked for
|
|
|
|
%s
|
|
|
|
Your current PATH is:
|
|
|
|
%s
|
|
|
|
We also looked for orted in the following directory:
|
|
|
|
%s
|
|
|
|
You may need to set your PATH properly, or set the MCA parameter
|
|
pls_bproc_orted to be the path to "orted".
|
|
[daemon-launch-number]
|
|
The bproc PLS component was not able to launch all the daemons on the remote
|
|
nodes and therefore cannot continue.
|
|
|
|
We attempted to launch %d daemons but only %d were actually launched.
|
|
|
|
For reference, we tried to launch %s
|
|
[daemon-launch-bad-pid]
|
|
The bproc PLS component was not able to launch all the daemons on the remote
|
|
nodes and therefore cannot continue.
|
|
|
|
On node %d the daemon pid was %d and errno was set to %d.
|
|
|
|
For reference, we tried to launch %s
|
|
[daemon-died-no-signal]
|
|
A daemon (pid %d) launched by the bproc PLS component on node %d died
|
|
unexpectedly so we are aborting.
|
|
|
|
This may be because the daemon was unable to find all the needed shared
|
|
libraries on the remote node. You may set your LD_LIBRARY_PATH to have the
|
|
location of the shared libraries on the remote nodes and this will
|
|
automatically be forwarded to the remote nodes.
|
|
[daemon-died-signal]
|
|
A daemon (pid %d) launched by the bproc PLS component on node %d died
|
|
unexpectedly on signal %d so we are aborting.
|
|
|
|
This may be because the daemon was unable to find all the needed shared
|
|
libraries on the remote node. You may set your LD_LIBRARY_PATH to have the
|
|
location of the shared libraries on the remote nodes and this will
|
|
automatically be forwarded to the remote nodes.
|
|
[proc-launch-number]
|
|
The bproc PLS component was not able to launch all the processes on the remote
|
|
nodes and therefore cannot continue.
|
|
|
|
We attempted to launch %d processes but only %d were actually launched.
|
|
|
|
For reference, we tried to launch %s
|
|
[proc-launch-bad-pid]
|
|
The bproc PLS component was not able to launch all the processes on the remote
|
|
nodes and therefore cannot continue.
|
|
|
|
On node %d the process pid was %d and errno was set to %d.
|
|
|
|
For reference, we tried to launch %s
|
|
|
|
[proc-io-setup-failed]
|
|
The daemons failed to setup the I/O forwarding subsystem to support
|
|
the application. This is a fatal error so we are aborting. For
|
|
reference, the internal error code was:
|
|
|
|
error: %s
|
|
|