Re: [OMPI users] Latencies of atomic operations on high-performance networks
Ok, then it sounds like a regression. I will try to track it down today or
tomorrow.
-Nathan
On Nov 08, 2018, at 01:41 PM, Joseph Schuchart wrote:
Sorry for the delay, I wanted to make sure that I test the same version
on both Aries and IB: git master bbe5da4. I realized that I had
previously tested with 3.1.3 on the IB cluster, which ran fine. If I use
the same version I run into the same problem on both systems (with --mca
btl_openib_allow_ib true --mca osc_rdma_acc_single_intrinsic true). I
have not tried using UCX for this.
Joseph
On 11/8/18 1:20 PM, Nathan Hjelm via users wrote:
Quick scan of the program and it looks ok to me. I will dig deeper and
see if I can determine the underlying cause.
What Open MPI version are you using?
-Nathan
On Nov 08, 2018, at 11:10 AM, Joseph Schuchart wrote:
While using the mca parameter in a real application I noticed a strange
effect, which took me a while to figure out: It appears that on the
Aries network the accumulate operations are not atomic anymore. I am
attaching a test program that shows the problem: all but one processes
continuously increment a counter while rank 0 is continuously
subtracting a large value and adding it again, eventually checking for
the correct number of increments. Without the mca parameter the test at
the end succeeds as all increments are accounted for:
```
$ mpirun -n 16 -N 1 ./mpi_fetch_op_local_remote
result:15000
```
When setting the mca parameter the test fails with garbage in the result:
```
$ mpirun --mca osc_rdma_acc_single_intrinsic true -n 16 -N 1
./mpi_fetch_op_local_remote
result:25769849013
mpi_fetch_op_local_remote: mpi_fetch_op_local_remote.c:97: main:
Assertion `sum == 1000*(comm_size-1)' failed.
```
All processes perform only MPI_Fetch_and_op in combination with MPI_SUM
so I assume that the test in combination with the mca flag is correct. I
cannot reproduce this issue on our IB cluster.
Is that an issue in Open MPI or is there some problem in the test case
that I am missing?
Thanks in advance,
Joseph
On 11/6/18 1:15 PM, Joseph Schuchart wrote:
Thanks a lot for the quick reply, setting
osc_rdma_acc_single_intrinsic=true does the trick for both shared and
exclusive locks and brings it down to <2us per operation. I hope that
the info key will make it into the next version of the standard, I
certainly have use for it :)
Cheers,
Joseph
On 11/6/18 12:13 PM, Nathan Hjelm via users wrote:
All of this is completely expected. Due to the requirements of the
standard it is difficult to make use of network atomics even for
MPI_Compare_and_swap (MPI_Accumulate and MPI_Get_accumulate spoil the
party). If you want MPI_Fetch_and_op to be fast set this MCA parameter:
osc_rdma_acc_single_intrinsic=true
Shared lock is slower than an exclusive lock because there is an extra
lock step as part of the accumulate (it isn't needed if there is an
exclusive lock). When setting the above parameter you are telling the
implementation that you will only be using a single count and we can
optimize that with the hardware. The RMA working group is working on
an info key that will essentially do the same thing.
Note the above parameter won't help you with IB if you are using UCX
unless you set this (master only right now):
btl_uct_transports=dc_mlx5
btl=self,vader,uct
osc=^ucx
Though there may be a way to get osc/ucx to enable the same sort of
optimization. I don't know.
-Nathan
On Nov 06, 2018, at 09:38 AM, Joseph Schuchart
wrote:
All,
I am currently experimenting with MPI atomic operations and wanted to
share some interesting results I am observing. The numbers below are
measurements from both an IB-based cluster and our Cray XC40. The
benchmarks look like the following snippet:
```
if (rank == 1) {
uint64_t res, val;
for (size_t i = 0; i < NUM_REPS; ++i) {
MPI_Fetch_and_op(&val, &res, MPI_UINT32_T, 0, 0, MPI_SUM, win);
MPI_Win_flush(target, win);
}
}
MPI_Barrier(MPI_COMM_WORLD);
```
Only rank 1 performs atomic operations, rank 0 waits in a barrier (I
have tried to confirm that the operations are done in hardware by
letting rank 0 sleep for a while and ensuring that communication
progresses). Of particular interest for my use-case is fetch_op but
I am
including other operations here nevertheless:
* Linux Cluster, IB QDR *
average of 10 iterations
Exclusive lock, MPI_UINT32_T:
fetch_op: 4.323384us
compare_exchange: 2.035905us
accumulate: 4.326358us
get_accumulate: 4.334831us
Exclusive lock, MPI_UINT64_T:
fetch_op: 2.438080us
compare_exchange: 2.398836us
accumulate: 2.435378us
get_accumulate: 2.448347us
Shared lock, MPI_UINT32_T:
fetch_op: 6.819977us
compare_exchange: 4.551417us
accumulate: 6.807766us
get_accumulate: 6.817602us
Shared lock, MPI_UINT64_T:
fetch_op: 4.954860us
compare_exchange: 2.399373us
accumulate: 4.965702us
get_accumulate: 4.977876us
There are two interesting observations:
a) operations on 64bit operands generally seem to have lower latencies
tha
Re: [OMPI users] Latencies of atomic operations on high-performance networks
Sorry for the delay, I wanted to make sure that I test the same version
on both Aries and IB: git master bbe5da4. I realized that I had
previously tested with 3.1.3 on the IB cluster, which ran fine. If I use
the same version I run into the same problem on both systems (with --mca
btl_openib_allow_ib true --mca osc_rdma_acc_single_intrinsic true). I
have not tried using UCX for this.
Joseph
On 11/8/18 1:20 PM, Nathan Hjelm via users wrote:
Quick scan of the program and it looks ok to me. I will dig deeper and
see if I can determine the underlying cause.
What Open MPI version are you using?
-Nathan
On Nov 08, 2018, at 11:10 AM, Joseph Schuchart wrote:
While using the mca parameter in a real application I noticed a strange
effect, which took me a while to figure out: It appears that on the
Aries network the accumulate operations are not atomic anymore. I am
attaching a test program that shows the problem: all but one processes
continuously increment a counter while rank 0 is continuously
subtracting a large value and adding it again, eventually checking for
the correct number of increments. Without the mca parameter the test at
the end succeeds as all increments are accounted for:
```
$ mpirun -n 16 -N 1 ./mpi_fetch_op_local_remote
result:15000
```
When setting the mca parameter the test fails with garbage in the result:
```
$ mpirun --mca osc_rdma_acc_single_intrinsic true -n 16 -N 1
./mpi_fetch_op_local_remote
result:25769849013
mpi_fetch_op_local_remote: mpi_fetch_op_local_remote.c:97: main:
Assertion `sum == 1000*(comm_size-1)' failed.
```
All processes perform only MPI_Fetch_and_op in combination with MPI_SUM
so I assume that the test in combination with the mca flag is correct. I
cannot reproduce this issue on our IB cluster.
Is that an issue in Open MPI or is there some problem in the test case
that I am missing?
Thanks in advance,
Joseph
On 11/6/18 1:15 PM, Joseph Schuchart wrote:
Thanks a lot for the quick reply, setting
osc_rdma_acc_single_intrinsic=true does the trick for both shared and
exclusive locks and brings it down to <2us per operation. I hope that
the info key will make it into the next version of the standard, I
certainly have use for it :)
Cheers,
Joseph
On 11/6/18 12:13 PM, Nathan Hjelm via users wrote:
All of this is completely expected. Due to the requirements of the
standard it is difficult to make use of network atomics even for
MPI_Compare_and_swap (MPI_Accumulate and MPI_Get_accumulate spoil the
party). If you want MPI_Fetch_and_op to be fast set this MCA parameter:
osc_rdma_acc_single_intrinsic=true
Shared lock is slower than an exclusive lock because there is an extra
lock step as part of the accumulate (it isn't needed if there is an
exclusive lock). When setting the above parameter you are telling the
implementation that you will only be using a single count and we can
optimize that with the hardware. The RMA working group is working on
an info key that will essentially do the same thing.
Note the above parameter won't help you with IB if you are using UCX
unless you set this (master only right now):
btl_uct_transports=dc_mlx5
btl=self,vader,uct
osc=^ucx
Though there may be a way to get osc/ucx to enable the same sort of
optimization. I don't know.
-Nathan
On Nov 06, 2018, at 09:38 AM, Joseph Schuchart
wrote:
All,
I am currently experimenting with MPI atomic operations and wanted to
share some interesting results I am observing. The numbers below are
measurements from both an IB-based cluster and our Cray XC40. The
benchmarks look like the following snippet:
```
if (rank == 1) {
uint64_t res, val;
for (size_t i = 0; i < NUM_REPS; ++i) {
MPI_Fetch_and_op(&val, &res, MPI_UINT32_T, 0, 0, MPI_SUM, win);
MPI_Win_flush(target, win);
}
}
MPI_Barrier(MPI_COMM_WORLD);
```
Only rank 1 performs atomic operations, rank 0 waits in a barrier (I
have tried to confirm that the operations are done in hardware by
letting rank 0 sleep for a while and ensuring that communication
progresses). Of particular interest for my use-case is fetch_op but
I am
including other operations here nevertheless:
* Linux Cluster, IB QDR *
average of 10 iterations
Exclusive lock, MPI_UINT32_T:
fetch_op: 4.323384us
compare_exchange: 2.035905us
accumulate: 4.326358us
get_accumulate: 4.334831us
Exclusive lock, MPI_UINT64_T:
fetch_op: 2.438080us
compare_exchange: 2.398836us
accumulate: 2.435378us
get_accumulate: 2.448347us
Shared lock, MPI_UINT32_T:
fetch_op: 6.819977us
compare_exchange: 4.551417us
accumulate: 6.807766us
get_accumulate: 6.817602us
Shared lock, MPI_UINT64_T:
fetch_op: 4.954860us
compare_exchange: 2.399373us
accumulate: 4.965702us
get_accumulate: 4.977876us
There are two interesting observations:
a) operations on 64bit operands generally seem to have lower latencies
than operations on 32bit
b) Using an exclusive lock leads to lower latencies
Overall, there is a factor of almost 3 between SharedLock+uint32_t and
ExclusiveLock+uint64_t for fetc
Re: [OMPI users] Latencies of atomic operations on high-performance networks
Quick scan of the program and it looks ok to me. I will dig deeper and see if I can determine the underlying cause.What Open MPI version are you using?-NathanOn Nov 08, 2018, at 11:10 AM, Joseph Schuchart wrote:While using the mca parameter in a real application I noticed a strange effect, which took me a while to figure out: It appears that on the Aries network the accumulate operations are not atomic anymore. I am attaching a test program that shows the problem: all but one processes continuously increment a counter while rank 0 is continuously subtracting a large value and adding it again, eventually checking for the correct number of increments. Without the mca parameter the test at the end succeeds as all increments are accounted for:```$ mpirun -n 16 -N 1 ./mpi_fetch_op_local_remoteresult:15000```When setting the mca parameter the test fails with garbage in the result:```$ mpirun --mca osc_rdma_acc_single_intrinsic true -n 16 -N 1 ./mpi_fetch_op_local_remoteresult:25769849013mpi_fetch_op_local_remote: mpi_fetch_op_local_remote.c:97: main: Assertion `sum == 1000*(comm_size-1)' failed.```All processes perform only MPI_Fetch_and_op in combination with MPI_SUM so I assume that the test in combination with the mca flag is correct. I cannot reproduce this issue on our IB cluster.Is that an issue in Open MPI or is there some problem in the test case that I am missing?Thanks in advance,JosephOn 11/6/18 1:15 PM, Joseph Schuchart wrote:Thanks a lot for the quick reply, settingosc_rdma_acc_single_intrinsic=true does the trick for both shared andexclusive locks and brings it down to <2us per operation. I hope thatthe info key will make it into the next version of the standard, Icertainly have use for it :)Cheers,JosephOn 11/6/18 12:13 PM, Nathan Hjelm via users wrote:All of this is completely expected. Due to the requirements of thestandard it is difficult to make use of network atomics even forMPI_Compare_and_swap (MPI_Accumulate and MPI_Get_accumulate spoil theparty). If you want MPI_Fetch_and_op to be fast set this MCA parameter:osc_rdma_acc_single_intrinsic=trueShared lock is slower than an exclusive lock because there is an extralock step as part of the accumulate (it isn't needed if there is anexclusive lock). When setting the above parameter you are telling theimplementation that you will only be using a single count and we canoptimize that with the hardware. The RMA working group is working onan info key that will essentially do the same thing.Note the above parameter won't help you with IB if you are using UCXunless you set this (master only right now):btl_uct_transports=dc_mlx5btl=self,vader,uctosc=^ucxThough there may be a way to get osc/ucx to enable the same sort ofoptimization. I don't know.-NathanOn Nov 06, 2018, at 09:38 AM, Joseph Schuchart wrote:All,I am currently experimenting with MPI atomic operations and wanted toshare some interesting results I am observing. The numbers below aremeasurements from both an IB-based cluster and our Cray XC40. Thebenchmarks look like the following snippet:```if (rank == 1) {uint64_t res, val;for (size_t i = 0; i < NUM_REPS; ++i) {MPI_Fetch_and_op(&val, &res, MPI_UINT32_T, 0, 0, MPI_SUM, win);MPI_Win_flush(target, win);}}MPI_Barrier(MPI_COMM_WORLD);```Only rank 1 performs atomic operations, rank 0 waits in a barrier (Ihave tried to confirm that the operations are done in hardware byletting rank 0 sleep for a while and ensuring that communicationprogresses). Of particular interest for my use-case is fetch_op but I amincluding other operations here nevertheless:* Linux Cluster, IB QDR *average of 10 iterationsExclusive lock, MPI_UINT32_T:fetch_op: 4.323384uscompare_exchange: 2.035905usaccumulate: 4.326358usget_accumulate: 4.334831usExclusive lock, MPI_UINT64_T:fetch_op: 2.438080uscompare_exchange: 2.398836usaccumulate: 2.435378usget_accumulate: 2.448347usShared lock, MPI_UINT32_T:fetch_op: 6.819977uscompare_exchange: 4.551417usaccumulate: 6.807766usget_accumulate: 6.817602usShared lock, MPI_UINT64_T:fetch_op: 4.954860uscompare_exchange: 2.399373usaccumulate: 4.965702usget_accumulate: 4.977876usThere are two interesting observations:a) operations on 64bit operands generally seem to have lower latenciesthan operations on 32bitb) Using an exclusive lock leads to lower latenciesOverall, there is a factor of almost 3 between SharedLock+uint32_t andExclusiveLock+uint64_t for fetch_and_op, accumulate, and get_accumulate(compare_exchange seems to be somewhat of an outlier).* Cray XC40, Aries *average of 10 iterationsExclusive lock, MPI_UINT32_T:fetch_op: 2.011794uscompare_exchange: 1.740825usaccumulate: 1.795500usget_accumulate: 1.985409usExclusive lock, MPI_UINT64_T:fetch_op: 2.017172uscompare_exchange: 1.846202usaccumulate: 1.812578usget_accumulate: 2.005541usShared lock, MPI_UINT32_T:fetch_op: 5.380455uscompare_exchange: 5.164458usaccumulate: 5.230184usget_accumulate: 5.399722usShared lock, MPI_UINT64_T:fetch_op: 5.415230uscompare_exchange: 1.855840usaccumul
Re: [OMPI users] Latencies of atomic operations on high-performance networks
While using the mca parameter in a real application I noticed a strange
effect, which took me a while to figure out: It appears that on the
Aries network the accumulate operations are not atomic anymore. I am
attaching a test program that shows the problem: all but one processes
continuously increment a counter while rank 0 is continuously
subtracting a large value and adding it again, eventually checking for
the correct number of increments. Without the mca parameter the test at
the end succeeds as all increments are accounted for:
```
$ mpirun -n 16 -N 1 ./mpi_fetch_op_local_remote
result:15000
```
When setting the mca parameter the test fails with garbage in the result:
```
$ mpirun --mca osc_rdma_acc_single_intrinsic true -n 16 -N 1
./mpi_fetch_op_local_remote
result:25769849013
mpi_fetch_op_local_remote: mpi_fetch_op_local_remote.c:97: main:
Assertion `sum == 1000*(comm_size-1)' failed.
```
All processes perform only MPI_Fetch_and_op in combination with MPI_SUM
so I assume that the test in combination with the mca flag is correct. I
cannot reproduce this issue on our IB cluster.
Is that an issue in Open MPI or is there some problem in the test case
that I am missing?
Thanks in advance,
Joseph
On 11/6/18 1:15 PM, Joseph Schuchart wrote:
Thanks a lot for the quick reply, setting
osc_rdma_acc_single_intrinsic=true does the trick for both shared and
exclusive locks and brings it down to <2us per operation. I hope that
the info key will make it into the next version of the standard, I
certainly have use for it :)
Cheers,
Joseph
On 11/6/18 12:13 PM, Nathan Hjelm via users wrote:
All of this is completely expected. Due to the requirements of the
standard it is difficult to make use of network atomics even for
MPI_Compare_and_swap (MPI_Accumulate and MPI_Get_accumulate spoil the
party). If you want MPI_Fetch_and_op to be fast set this MCA parameter:
osc_rdma_acc_single_intrinsic=true
Shared lock is slower than an exclusive lock because there is an extra
lock step as part of the accumulate (it isn't needed if there is an
exclusive lock). When setting the above parameter you are telling the
implementation that you will only be using a single count and we can
optimize that with the hardware. The RMA working group is working on
an info key that will essentially do the same thing.
Note the above parameter won't help you with IB if you are using UCX
unless you set this (master only right now):
btl_uct_transports=dc_mlx5
btl=self,vader,uct
osc=^ucx
Though there may be a way to get osc/ucx to enable the same sort of
optimization. I don't know.
-Nathan
On Nov 06, 2018, at 09:38 AM, Joseph Schuchart wrote:
All,
I am currently experimenting with MPI atomic operations and wanted to
share some interesting results I am observing. The numbers below are
measurements from both an IB-based cluster and our Cray XC40. The
benchmarks look like the following snippet:
```
if (rank == 1) {
uint64_t res, val;
for (size_t i = 0; i < NUM_REPS; ++i) {
MPI_Fetch_and_op(&val, &res, MPI_UINT32_T, 0, 0, MPI_SUM, win);
MPI_Win_flush(target, win);
}
}
MPI_Barrier(MPI_COMM_WORLD);
```
Only rank 1 performs atomic operations, rank 0 waits in a barrier (I
have tried to confirm that the operations are done in hardware by
letting rank 0 sleep for a while and ensuring that communication
progresses). Of particular interest for my use-case is fetch_op but I am
including other operations here nevertheless:
* Linux Cluster, IB QDR *
average of 10 iterations
Exclusive lock, MPI_UINT32_T:
fetch_op: 4.323384us
compare_exchange: 2.035905us
accumulate: 4.326358us
get_accumulate: 4.334831us
Exclusive lock, MPI_UINT64_T:
fetch_op: 2.438080us
compare_exchange: 2.398836us
accumulate: 2.435378us
get_accumulate: 2.448347us
Shared lock, MPI_UINT32_T:
fetch_op: 6.819977us
compare_exchange: 4.551417us
accumulate: 6.807766us
get_accumulate: 6.817602us
Shared lock, MPI_UINT64_T:
fetch_op: 4.954860us
compare_exchange: 2.399373us
accumulate: 4.965702us
get_accumulate: 4.977876us
There are two interesting observations:
a) operations on 64bit operands generally seem to have lower latencies
than operations on 32bit
b) Using an exclusive lock leads to lower latencies
Overall, there is a factor of almost 3 between SharedLock+uint32_t and
ExclusiveLock+uint64_t for fetch_and_op, accumulate, and get_accumulate
(compare_exchange seems to be somewhat of an outlier).
* Cray XC40, Aries *
average of 10 iterations
Exclusive lock, MPI_UINT32_T:
fetch_op: 2.011794us
compare_exchange: 1.740825us
accumulate: 1.795500us
get_accumulate: 1.985409us
Exclusive lock, MPI_UINT64_T:
fetch_op: 2.017172us
compare_exchange: 1.846202us
accumulate: 1.812578us
get_accumulate: 2.005541us
Shared lock, MPI_UINT32_T:
fetch_op: 5.380455us
compare_exchange: 5.164458us
accumulate: 5.230184us
get_accumulate: 5.399722us
Shared lock, MPI_UINT64_T:
fetch_op: 5.415230us
compare_exchange: 1.855840us
accumulate: 5.212632us
get_ac
Re: [OMPI users] Bug with Open-MPI Processor Count
Hello Ralph,
Is there any update on this?
Thanks,
Adam LeBlanc
On Fri, Nov 2, 2018 at 11:06 AM Adam LeBlanc wrote:
> Hello Ralph,
>
> When I do the -np 7 it still fails with "There are not enough slots
> available in the system to satisfy the 7 slots that were requested by the
> application", but when I do -np 2 it will actually run from a machine that
> was failing but will only run on one other machine and in this case it ran
> from a machine with 2 processors to a machine with only 1 processor. If I
> try to make -np higher then 2 it will also fail.
>
> -Adam LeBlanc
>
> On Thu, Nov 1, 2018 at 3:53 PM Ralph H Castain wrote:
>
>> Hmmm - try adding a value for nprocs instead of leaving it blank. Say
>> “-np 7”
>>
>> Sent from my iPhone
>>
>> On Nov 1, 2018, at 11:56 AM, Adam LeBlanc wrote:
>>
>> Hello Ralph,
>>
>> Here is the output for a failing machine:
>>
>> [130_02:44:13_aleblanc@farbauti]{~}$ > mpirun --mca
>> btl_openib_warn_no_device_params_found 0 --mca orte_base_help_aggregate 0
>> --mca btl openib,vader,self --mca pml ob1 --mca btl_openib_receive_queues
>> P,65536,120,64,32 -hostfile /home/soesterreich/ce-mpi-hosts --mca
>> ras_base_verbose 5 IMB-MPI1
>>
>> == ALLOCATED NODES ==
>> farbauti: flags=0x11 slots=1 max_slots=0 slots_inuse=0 state=UP
>> hyperion-ce: flags=0x10 slots=1 max_slots=0 slots_inuse=0 state=UNKNOWN
>> io-ce: flags=0x10 slots=1 max_slots=0 slots_inuse=0 state=UNKNOWN
>> jarnsaxa-ce: flags=0x10 slots=1 max_slots=0 slots_inuse=0 state=UNKNOWN
>> rhea-ce: flags=0x10 slots=1 max_slots=0 slots_inuse=0 state=UNKNOWN
>> tarqeq-ce: flags=0x10 slots=1 max_slots=0 slots_inuse=0 state=UNKNOWN
>> tarvos-ce: flags=0x10 slots=1 max_slots=0 slots_inuse=0 state=UNKNOWN
>> =
>> --
>> There are not enough slots available in the system to satisfy the 7 slots
>> that were requested by the application:
>> 10
>>
>> Either request fewer slots for your application, or make more slots
>> available
>> for use.
>> --
>>
>>
>> Here is an output of a passing machine:
>>
>> [1_02:54:26_aleblanc@hyperion]{~}$ > mpirun --mca
>> btl_openib_warn_no_device_params_found 0 --mca orte_base_help_aggregate 0
>> --mca btl openib,vader,self --mca pml ob1 --mca btl_openib_receive_queues
>> P,65536,120,64,32 -hostfile /home/soesterreich/ce-mpi-hosts --mca
>> ras_base_verbose 5 IMB-MPI1
>>
>> == ALLOCATED NODES ==
>> hyperion: flags=0x11 slots=1 max_slots=0 slots_inuse=0 state=UP
>> farbauti-ce: flags=0x10 slots=1 max_slots=0 slots_inuse=0 state=UNKNOWN
>> io-ce: flags=0x10 slots=1 max_slots=0 slots_inuse=0 state=UNKNOWN
>> jarnsaxa-ce: flags=0x10 slots=1 max_slots=0 slots_inuse=0 state=UNKNOWN
>> rhea-ce: flags=0x10 slots=1 max_slots=0 slots_inuse=0 state=UNKNOWN
>> tarqeq-ce: flags=0x10 slots=1 max_slots=0 slots_inuse=0 state=UNKNOWN
>> tarvos-ce: flags=0x10 slots=1 max_slots=0 slots_inuse=0 state=UNKNOWN
>> =
>>
>>
>> Yes the hostfile is available on all nodes through an NFS mount for all
>> of our home directories.
>>
>> On Thu, Nov 1, 2018 at 2:44 PM Adam LeBlanc wrote:
>>
>>>
>>>
>>> -- Forwarded message -
>>> From: Ralph H Castain
>>> Date: Thu, Nov 1, 2018 at 2:34 PM
>>> Subject: Re: [OMPI users] Bug with Open-MPI Processor Count
>>> To: Open MPI Users
>>>
>>>
>>> I’m a little under the weather and so will only be able to help a bit at
>>> a time. However, a couple of things to check:
>>>
>>> * add -mca ras_base_verbose 5 to the cmd line to see what mpirun thought
>>> the allocation was
>>>
>>> * is the hostfile available on every node?
>>>
>>> Ralph
>>>
>>> On Nov 1, 2018, at 10:55 AM, Adam LeBlanc wrote:
>>>
>>> Hello Ralph,
>>>
>>> Attached below is the verbose output for a failing machine and a passing
>>> machine.
>>>
>>> Thanks,
>>> Adam LeBlanc
>>>
>>> On Thu, Nov 1, 2018 at 1:41 PM Adam LeBlanc
>>> wrote:
>>>
-- Forwarded message -
From: Ralph H Castain
Date: Thu, Nov 1, 2018 at 1:07 PM
Subject: Re: [OMPI users] Bug with Open-MPI Processor Count
To: Open MPI Users
Set rmaps_base_verbose=10 for debugging output
Sent from my iPhone
On Nov 1, 2018, at 9:31 AM, Adam LeBlanc wrote:
The version by the way for Open-MPI is 3.1.2.
-Adam LeBlanc
On Thu, Nov 1, 2018 at 12:05 PM Adam LeBlanc
wrote:
> Hello, I am an employee of the UNH InterOperability Lab, and we are in
> the process of testing OFED-4.17-RC1 for the OpenFabrics Alliance. We have
> purchased some new hardware that has one processor, and noticed an issue
> when running mpi jobs on nodes that do not have similar processor count
