[Wien] Help with Operation System

[Peter Blaha]

Thanks for the benchmark times. This seems to be a nice processor.
Unfortunately, some of them are not very clear.

a) Please set:
export OMP_NUM_THREADS=1 (2 and 4)
run    x lapw1
and check the timings (it should go faster when using more threads). You can also
verify using the "top" command whether the lapw1 command uses 1,2 or 4 cores.

b) x lapw1 -p    is not enough. What you should do is:
    create 4 identical directories, then open 4 windows, change into these 4 directories and
    run  "x lapw1" at the same time 2 or 4 times and check the timings.

c) Most likely, the mpi-benchmark is not using mpi,.... But forget this test.

Am 17.06.2011 05:01, schrieb M. Koshino:
> Dear all,
>
> In my opinion, xeon is not a new cpu so that it does not matter what OS you choose as long as your system and compilers properly recognize your hardware.
>
> I recently set up a very cheap, new PC by myself and here is a note to encourage researchers of less funded.
>
> When I first installed CentOS 5.6 on my core i7 2600 machine, the cpu was not properly recognized (there could be some ways but I easily gave up).
> The intel compilers and other software were therefore not properly installed on the system with core i7 2600.
>
> I could set up the latest intel compilers properly on fedora core 15, and could compile wien2k ver.11 (properly I hope).
>
> System:
> CPU: Core i7 2600 BOX
> (recognized as processor 0-7: core i7-2600 cpu @ 3.40 GHz)
> Motherboard: ASUS, P8P67 DELUXE REV 3.0
> Video Card: ASUS, EAH5850 DIRECTCU/2DIS/1GD5 (PCIExp 1GB)
> OS: Linux 2.6.38.6-27.fc15.x86_64
> GNOME 3.0.1
> memory: 16 GB
> Corsair CMZ8GX3M2A1600C8 (4GBx4)
> compiler
> /opt/intel/composerxe-2011.4.191/bin/intel64/ifort
> applying update 4.0 for C++ compiler XE 12.0, Fortran Compiler XE 12.0, Debugger 12.0, MKL 10.3, IPP 7.0, TBB 3.0.
> (I could save $20K by comparing to my previous purchase from a high performance computer vendor.)
>
> After running the test_case
> command "x lapw1 -c"
> test_case.output1:       TIME HAMILT (CPU)  =     6.5, HNS =     4.4, HORB =     0.0, DIAG =    25.4
> test_case.output1:       TIME HAMILT (WALL) =     6.9, HNS =     4.4, HORB =     0.0, DIAG =    25.8
>
> command "x lapw1 -p"
> test_case.output1:       TIME HAMILT (CPU)  =     6.5, HNS =     4.4, HORB =     0.0, DIAG =    25.4
> test_case.output1:       TIME HAMILT (WALL) =     6.7, HNS =     4.4, HORB =     0.0, DIAG =    25.6
>
> after running the mpi-benchmark
> command "x lapw1 -p"
> Intel_core2_2.4_single_ifort-10_mkl-9.1.output1:       TIME HAMILT (CPU)  =   200.5, HNS =   154.5, HORB =     0.0, DIAG =   923.7
> mpi-benchmark.output1:       TIME HAMILT (CPU)  =    91.0, HNS =    50.6, HORB =     0.0, DIAG =   300.0
> mpi-benchmark.output1:       TIME HAMILT (WALL) =    91.8, HNS =    51.0, HORB =     0.0, DIAG =   303.4
>
> I am not familiar with the values here so I hope someone will make a comment on this.
>
> Kind regards,
>
> Masanori Koshino
>
> Nanotube Research Center
> AIST, JAPAN
> m-koshino at aist.go.jp
>
>
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien

-- 

                                       P.Blaha
--------------------------------------------------------------------------
Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-1-58801-15671             FAX: +43-1-58801-15698
Email: blaha at theochem.tuwien.ac.at    WWW: http://info.tuwien.ac.at/theochem/
--------------------------------------------------------------------------