On 12/6/07 8:06 AM, "Shipman, Galen M." <gship...@ornl.gov> wrote:
>>>
>>> Do we really need a complete node map? A far as I can tell, it looks
>>> like the MPI layer only needs a list of local processes. So maybe it
>>> would be better to forget about the node ids at the mpi layer and just
>>> return the local procs.
>>
>> I agree, though I don't think we want a parallel list of procs. We just need
>> to set the "local" flag in the existing ompi_proc_t structures.
>>
>
> Having a parallel list of procs makes perfect sense. That way ORTE can store
> ORTE information in the orte_proc_t and OMPI can store OMPI information in
> the ompi_proc_t. The ompi_proc_t could either "inherit" the orte_proc_t or
> have a pointer to it so that we have no duplication of data.
>
Hmmm....well, I personally don't have an opinion either way regarding how
the info flows up to the MPI layer, so I'll leave that up to you MPI folks.
I can certainly create such a list if that's the way you want to go.
However, since I have no way of knowing that a job is an MPI-job or not, I
will point out that non-MPI jobs will see an increased memory footprint as a
result. Not sure we care a whole lot - depends upon what other uses people
are trying to make of ORTE (e.g., STCI).
I will point out that getting the complete node map to every process will
incur some penalty in terms of launch time, and still begs the question of
how it gets there. I can provide it when launching via the ORTE daemons, of
course (would have to send it via a daemon-to-local-proc RML message,
though, as it would be too large to put in enviro variables) - I assume
there must be a parallel mechanism for something like the Cray to provide
it?
I'm also not sure how something like SLURM or TM would provide this if/when
we tightly integrate (i.e., go through their daemons instead of ORTE's) -
could be a future issue.
> Having a global map makes sense, particularly for numerous communication
> scenarios, if I know all the processes are on the same node I may send a
> message to the lowest "vpid" on that node and he could then forward to
> everyone else.
Are you talking about RML communications? If so, we already have that in
place via the daemons.
Or are you talking about routing BTL/MPI messages?? I thought that was a
"no-no"?
>
>
>> One option is for the RTE to just pass in an enviro variable with a
>> comma-separated list of your local ranks, but that creates a problem down
>> the road when trying to integrate tighter with systems like SLURM where the
>> procs would get mass-launched (so the environment has to be the same for all
>> of them).
>>
> Having a enviro variable with at comma-seperated list of local ranks doesn't
> seems like a bit of a hack to me.
>
>>>
>>> So my vote would be to leave the modex alone, but remove the node id,
>>> and add a function to get the list of local procs. It doesn't matter to
>>> me how the RTE implements that.
>>
>> I think we would need to be careful here that we don't create a need for
>> more communication. We have two functions currently in the modex:
>>
>> 1. how to exchange the info required to populate the ompi_proc_t structures;
>> and
>>
>> 2. how to identify which of those procs are "local"
>>
>> The problem with leaving the modex as it currently sits is that some
>> environments require a different mechanism for exchanging the ompi_proc_t
>> info. While most can use the RML, some can't. The same division of
>> capabilities applies to getting the "local" info, so it makes sense to me to
>> put the modex in a framework.
>>
>> Otherwise, we wind up with a bunch of #if's in the code to support
>> environments like the Cray. I believe the mca system was put in place
>> precisely to avoid those kind of practices, so it makes sense to me to take
>> advantage of it.
>>
>>
>>>
>>> Alternatively, if we did a process attribute system we could just use
>>> predefined attributes, and the runtime can get each process's node id
>>> however it wants.
>>
>> Same problem as above, isn't it? Probably ignorance on my part, but it seems
>> to me that we simply exchange a modex framework for an attribute framework
>> (since each environment would have to get the attribute values in a
>> different manner) - don't we?
>>
>> I have no problem with using attributes instead of the modex, but the issue
>> appears to be the same either way - you still need a framework to handle the
>> different methods.
>>
>>
>> Ralph
>>
>>>
>>> Tim
>>>
>>> Ralph H Castain wrote:
>>>> IV. RTE/MPI relative modex responsibilities
>>>> The modex operation conducted during MPI_Init currently involves the
>>>> exchange of two critical pieces of information:
>>>>
>>>> 1. the location (i.e., node) of each process in my job so I can determine
>>>> who shares a node with me. This is subsequently used by the shared memory
>>>> subsystem for initialization and message routing; and
>>>>
>>>> 2. BTL contact info for each process in my job.
>>>>
>>>> During our recent efforts to further abstract the RTE from the MPI layer,
>>>> we
>>>> pushed responsibility for both pieces of information into the MPI layer.
>>>> This wasn't done capriciously - the modex has always included the exchange
>>>> of both pieces of information, and we chose not to disturb that situation.
>>>>
>>>> However, the mixing of these two functional requirements does cause
>>>> problems
>>>> when dealing with an environment such as the Cray where BTL information is
>>>> "exchanged" via an entirely different mechanism. In addition, it has been
>>>> noted that the RTE (and not the MPI layer) actually "knows" the node
>>>> location for each process.
>>>>
>>>> Hence, questions have been raised as to whether:
>>>>
>>>> (a) the modex should be built into a framework to allow multiple BTL
>>>> exchange mechansims to be supported, or some alternative mechanism be used
>>>> -
>>>> one suggestion made was to implement an MPICH-like attribute exchange; and
>>>>
>>>> (b) the RTE should absorb responsibility for providing a "node map" of the
>>>> processes in a job (note: the modex may -use- that info, but would no
>>>> longer
>>>> be required to exchange it). This has a number of implications that need to
>>>> be carefully considered - e.g., the memory required to store the node map
>>>> in
>>>> every process is non-zero. On the other hand:
>>>>
>>>> (i) every proc already -does- store the node for every proc - it is simply
>>>> stored in the ompi_proc_t structures as opposed to somewhere in the RTE. We
>>>> would want to avoid duplicating that storage, but there would be no change
>>>> in memory footprint if done carefully.
>>>>
>>>> (ii) every daemon already knows the node map for the job, so communicating
>>>> that info to its local procs may not prove a major burden. However, the
>>>> very
>>>> environments where this subject may be an issue (e.g., the Cray) do not use
>>>> our daemons, so some alternative mechanism for obtaining the info would be
>>>> required.
>>>>
>>>>
>>>> So the questions to be considered here are:
>>>>
>>>> (a) do we leave the current modex "as-is", to include exchange of the node
>>>> map, perhaps including "#if" statements to support different exchange
>>>> mechanisms?
>>>>
>>>> (b) do we separate the two functions currently in the modex and push the
>>>> requirement to obtain a node map into the RTE? If so, how do we want the
>>>> MPI
>>>> layer to retrieve that info so we avoid increasing our memory footprint?
>>>>
>>>> (c) do we create a separate modex framework for handling the different
>>>> exchange mechanisms for BTL info, do we incorporate it into an existing one
>>>> (if so, which one), the new publish-subscribe framework, implement an
>>>> alternative approach, or...?
>>>>
>>>> (d) other suggestions?
>>>>
>>>> Ralph
>>>>
>>>>
>>>> _______________________________________________
>>>> devel mailing list
>>>> de...@open-mpi.org
>>>> http://www.open-mpi.org/mailman/listinfo.cgi/devel
>>>
>>
>>
>> _______________________________________________
>> devel mailing list
>> de...@open-mpi.org
>> http://www.open-mpi.org/mailman/listinfo.cgi/devel
>
