<div>Hi,</div><div><br></div><div>I have a query regarding Lustre Throughput Numbers with mdtest benchmark.I am running mdtest benhmark with following options :-</div><div><br></div><div>/home/meshram/mpich2-new/mpich2-1.4/mpich2-install/bin/mpirun -np 256 -hostfile ./hostfile ./mdtest -z 3 -b 10 -I 5 -v -d /tmp/l66</div>
<div><br></div><div><br></div><div>where ,</div><div>mdtest - is the standard benchmark to test metadata operations. [<a href="https://computing.llnl.gov/?set=code&page=sio_downloads">https://computing.llnl.gov/?set=code&page=sio_downloads</a>]</div>
<div>/tmp/l66 is my Lustre mount.</div><div>I am using 1Gige Network with TCP transport.</div><div>hostfile has 8 host nodes</div><div>I am varying the number of processes as we can see in following table .</div><div>I was amazed by the throughput which I got. I think this is too huge. Can someone please let me know if these numbers are correct ?</div>
<div><br></div><div><br></div><div><br></div>4 Process : File creation : 0.501 sec, 44392.140 ops/sec<br><div>8 Process :File creation : 0.685 sec, 64890.598 ops/sec</div><div>16 Process :File creation : 1.426 sec, 62318.798 ops/sec</div>
<div>32 Process :File creation : 2.947 sec, 60312.766 ops/sec</div><div>64 Process :File creation : 5.630 sec, 63142.760 ops/sec</div><div>128 Process :File creation : 13.208 sec, 53835.707 ops/sec</div>
<div>256 Process :File creation : 24.601 sec, 57804.777 ops/sec</div><div><br></div><div>Thanks,</div><div>Vilobh</div><div>Graduate Research Associate</div><div>Department of Computer Science</div><div>The Ohio State University Columbus Ohio</div>