This is a prime example how one can benefit from using the proper software, much more than when buying a more expensive hardware.

The speedup from 150 to 30 seconds (wall-time factor 5) on a 4-core cpu is impressive.

Peter Blaha

Am 24.11.2021 um 13:55 schrieb David Holec:
Dear Pavel,

Many thanks again for your patience and guidance. With the libopenblas-openmp-dev package it seems to work well!

$ ldd lapw1
       linux-vdso.so.1 (0x00007ffca83d8000)
*libopenblas.so.0 => /lib/x86_64-linux-gnu/libopenblas.so.0 (0x000014563f924000) *        libgfortran.so.5 => /lib/x86_64-linux-gnu/libgfortran.so.5 (0x000014563f65c000)
       libm.so.6 => /lib/x86_64-linux-gnu/libm.so.6 (0x000014563f50d000)
       libmvec.so.1 => /lib/x86_64-linux-gnu/libmvec.so.1 (0x000014563f4e1000)        libgomp.so.1 => /lib/x86_64-linux-gnu/libgomp.so.1 (0x000014563f49f000)        libpthread.so.0 => /lib/x86_64-linux-gnu/libpthread.so.0 (0x000014563f47c000)
       libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x000014563f288000)
       /lib64/ld-linux-x86-64.so.2 (0x0000145641b2d000)
       libquadmath.so.0 => /lib/x86_64-linux-gnu/libquadmath.so.0 (0x000014563f23e000)        libgcc_s.so.1 => /lib/x86_64-linux-gnu/libgcc_s.so.1 (0x000014563f223000)
       libdl.so.2 => /lib/x86_64-linux-gnu/libdl.so.2 (0x000014563f21d000)

and these options in siteconfig:
 L   Linker Flags:            $(FOPT) -L/usr/lib/x86_64-linux-gnu/openblas-openmp
 R   R_LIBS (LAPACK+BLAS):    -lopenblas

I also get better timings now (though the TIME HAMILT are slightly longer, but overall improvement):
$ ../x lapw1
STOP  LAPW1 END
119.400u 1.937s 0:32.53 372.9%  0+0k 0+37864io 0pf+0w
$ grep HORB *output1*
test_case.output1:      TIME HAMILT (CPU)  =    17.3, HNS =    18.4, HORB=     0.0, DIAG =    85.0, SYNC =     0.0 test_case.output1:      TIME HAMILT (WALL) =     4.6, HNS =     5.2, HORB=     0.0, DIAG =    22.0, SYNC =     0.0


> $ grep HORB *output1*

    > test_case.output1:       TIME HAMILT (CPU)  =    16.3, HNS =
       20.0,
    > HORB =     0.0, DIAG =   125.9, SYNC =     0.0
    > test_case.output1:       TIME HAMILT (WALL) =     4.4, HNS =
       20.0,
    > HORB =     0.0, DIAG =   126.0, SYNC =     0.0

--
-----------------------------------------------------------------------
Peter Blaha,  Inst. f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-158801165300
Email:peter.bl...@tuwien.ac.at WWW:http://www.imc.tuwien.ac.at WIEN2k:http://www.wien2k.at
-------------------------------------------------------------------------
_______________________________________________
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html

Reply via email to