Open MPI logo

Open MPI User's Mailing List Archives

  |   Home   |   Support   |   FAQ   |   all Open MPI User's mailing list

Subject: Re: [OMPI users] Myrinet optimization with OMP1.3 and macosX
From: Ricardo Fernández-Perea (rfernandezperea_at_[hidden])
Date: 2009-03-20 11:33:10


On Fri, Mar 20, 2009 at 2:21 PM, Scott Atchley <atchley_at_[hidden]> wrote:

> On Mar 20, 2009, at 5:59 AM, Ricardo Fernández-Perea wrote:
>
> Hello,
>>
>> I am running DL_POLY in various Xserver 8 processor with a myrinet
>> network.using mx-1.2.7
>>
>> While I keep in the same node the process scales reasonably well but in
>> the moment I hit the network ...
>>
>> I will like to try to maximize the mx network before trying to touch the
>> program code.
>>
>
> Have you run mx_pingpong between the nodes to determine what raw
> performance you are getting?
>
> For each pair of machines, run:
>
> Host foo:
> $ mx_pingpong -s
>
> Host bar:
> $ mx_pingpong -d foo -S 0 -E 5000000 -M 2
>
> where "-d foo" means the MX hostname of foo. It could be foo or foo:0 (or
> foo:1 if you have two NICs, etc.).
>
> See http://www.myri.com/serve/cache/455.html for more details.
>
> Repeat the above after setting MX_RCACHE=1 on both hosts.

This are the results
initially
Running 1000 iterations.
   Length Latency(us) Bandwidth(MB/s)
        0 2.738 0.000
        1 2.718 0.368
        2 2.707 0.739
        4 2.715 1.474
        8 2.719 2.943
       16 2.748 5.822
       32 2.814 11.370
       64 3.333 19.199
      128 3.808 33.609
      256 6.069 42.185
      512 8.101 63.206
     1024 11.841 86.483
     2048 16.100 127.205
     4096 24.942 164.221
     8192 44.332 184.788
    16384 83.784 195.550
    32768 150.186 218.182
    65536 318.591 205.706
   131072 591.277 221.676
   262144 1128.918 232.208
   524288 2215.368 236.660
  1048576 4392.217 238.735
  2097152 8705.028 240.913
  4194304 17359.166 241.619

with export MX_RCACHE=1

Running 1000 iterations.
   Length Latency(us) Bandwidth(MB/s)
        0 2.731 0.000
        1 2.705 0.370
        2 2.719 0.736
        4 2.703 1.480
        8 2.709 2.953
       16 2.743 5.833
       32 2.811 11.386
       64 3.336 19.188
      128 3.815 33.552
      256 6.029 42.465
      512 8.082 63.347
     1024 11.814 86.677
     2048 16.113 127.098
     4096 24.945 164.205
     8192 44.354 184.696
    16384 83.769 195.587
    32768 150.183 218.187
    65536 299.793 218.604
   131072 561.008 233.637
   262144 1093.738 239.677
   524288 2150.151 243.838
  1048576 4265.846 245.807
  2097152 8491.122 246.982
  4194304 16953.997 247.393

Ricardo

>
>
> Scott
>
> Is there some place where I can check what the mca params really means?
>> Can anyone give me a hint where it would be better to look to optimize
>> OMPI1.3 for macosx?
>>
>> Thanks
>>
>> Ricardo
>> _______________________________________________
>> users mailing list
>> users_at_[hidden]
>> http://www.open-mpi.org/mailman/listinfo.cgi/users
>>
>
>
> _______________________________________________
> users mailing list
> users_at_[hidden]
> http://www.open-mpi.org/mailman/listinfo.cgi/users
>