1
1
openmpi/opal/mca/hwloc
Ralph Castain 6310361532 At long last, the fabled revision to the affinity system has arrived. A more detailed explanation of how this all works will be presented here:
https://svn.open-mpi.org/trac/ompi/wiki/ProcessPlacement

The wiki page is incomplete at the moment, but I hope to complete it over the next few days. I will provide updates on the devel list. As the wiki page states, the default and most commonly used options remain unchanged (except as noted below). New, esoteric and complex options have been added, but unless you are a true masochist, you are unlikely to use many of them beyond perhaps an initial curiosity-motivated experimentation.

In a nutshell, this commit revamps the map/rank/bind procedure to take into account topology info on the compute nodes. I have, for the most part, preserved the default behaviors, with three notable exceptions:

1. I have at long last bowed my head in submission to the system admin's of managed clusters. For years, they have complained about our default of allowing users to oversubscribe nodes - i.e., to run more processes on a node than allocated slots. Accordingly, I have modified the default behavior: if you are running off of hostfile/dash-host allocated nodes, then the default is to allow oversubscription. If you are running off of RM-allocated nodes, then the default is to NOT allow oversubscription. Flags to override these behaviors are provided, so this only affects the default behavior.

2. both cpus/rank and stride have been removed. The latter was demanded by those who didn't understand the purpose behind it - and I agreed as the users who requested it are no longer using it. The former was removed temporarily pending implementation.

3. vm launch is now the sole method for starting OMPI. It was just too darned hard to maintain multiple launch procedures - maybe someday, provided someone can demonstrate a reason to do so.

As Jeff stated, it is impossible to fully test a change of this size. I have tested it on Linux and Mac, covering all the default and simple options, singletons, and comm_spawn. That said, I'm sure others will find problems, so I'll be watching MTT results until this stabilizes.

This commit was SVN r25476.
2011-11-15 03:40:11 +00:00
..
base At long last, the fabled revision to the affinity system has arrived. A more detailed explanation of how this all works will be presented here: 2011-11-15 03:40:11 +00:00
external Fix some problems with the libevent and hwloc frameworks: 2011-09-21 23:06:13 +00:00
hwloc122ompi At long last, the fabled revision to the affinity system has arrived. A more detailed explanation of how this all works will be presented here: 2011-11-15 03:40:11 +00:00
configure.m4 We always have hwloc xml support (now that it's built into to hwloc 2011-10-11 20:20:59 +00:00
hwloc.h At long last, the fabled revision to the affinity system has arrived. A more detailed explanation of how this all works will be presented here: 2011-11-15 03:40:11 +00:00
Makefile.am Per the RFC from Jeff, move hwloc from opal/mca/common to its own static framework ala libevent. Have ORTE daemons collect the topology info at startup and, if --enable-hwloc-xml is set, send that info back to the HNP for later use. The HNP only retains unique topology "templates" to reduce memory footprint. Have the daemon include the local topology info in the nidmap buffer sent to each app so the apps don't all hammer the local system to discover it for themselves. 2011-09-11 19:02:24 +00:00
README.txt Add some notes about maintaining the hwloc framework. 2011-09-12 19:40:18 +00:00

12 Sep 2011

Notes for hwloc component maintainers:

1. There can only be *1* hwloc version component at a time.
   Specifically: if there are multiple hwlocXYZ components (i.e.,
   different versions of hwloc), then they must all be .ompi_ignore'd
   except for 1.  This is because we currently m4_include all of the
   underlying hwloc's .m4 files -- if there are multiple hwlocXYZ
   components, I don't know if m4 will barf at the multiple,
   conflicting AC_DEFUNs, or whether it'll just do something
   completely undefined.

1a. As a consequence, if you're adding a new hwloc version component,
   you'll need to .ompi_ignore all others while you're testing the new
   one. 

2. If someone wants to fix #1 someday, we might be able to do what we
   do for libevent: OMPI_CONFIG_SUBDIR (instead of slurping in hwloc's
   .m4 files).