Well ... this is exactly the kind of behavior a high performance
application try to achieve isn't it ?
The problem here is not the flow control. What you need is to avoid
buffering the messages on the receiver side. Luckily, Open MPI is
entirely configurable at runtime, so this situation is really easy to
deal with even at the user level. Set the eager size to zero, and no
buffering on the receiver side will be made. Your program will survive
as long as there is some available memory on the receiver.
On Feb 1, 2008, at 6:32 PM, 8mj6tc902_at_[hidden] wrote:
> That would make sense. I able to break OpenMPI by having Node A wait
> messages from Node B. Node B is in fact sleeping while Node C bombards
> Node A with a few thousand messages. After a while Node B wakes up and
> sends Node A the message it's been waiting on, but Node A has long
> been buried and seg faults. If I decrease the number of messages C is
> sending, it works properly. This was on OpenMPI 1.2.4 (using I think
> SM BTL (might have been MX or TCP, but certainly not infiniband. I
> dig up the test and try again if anyone is seriously curious).
> Trying the same test on MPICH/MX went very very slow (I don't think
> have any clever buffer management) but it didn't crash.
> Sacerdoti, Federico Federico.Sacerdoti-at-deshaw.com
> |openmpi-users/Allow| wrote:
>> I am readying an openmpi 1.2.5 software stack for use with a
>> many-thousand core cluster. I have a question about sending small
>> messages that I hope can be answered on this list.
>> I was under the impression that if node A wants to send a small MPI
>> message to node B, it must have a credit to do so. The credit
>> assures A
>> that B has enough buffer space to accept the message. Credits are
>> required by the mpi layer regardless of the BTL transport layer used.
>> I have been told by a Voltaire tech that this is not so, the
>> credits are
>> used by the infiniband transport layer to reliably send a message,
>> is not an openmpi feature.
>> users mailing list
> [A dream that comes true can't really be called a dream.]
> users mailing list
- application/pkcs7-signature attachment: smime.p7s