[Wien] (no subject)

Mon Nov 3 05:49:34 CET 2003

Dear peter
    I have calculated a big case,which have 49 atoms,have Zn ,N,S,O,C,and H.
After succesful iteration,it took about 2800 minutes,for the second
iteration,when run lapw2,it went wrong,then stop.
dayfile:
   start (三 10月 29 19:14:48 CST 2003) with lapw0 (20/20 to go)
  lapw0 -p (19:14:48) starting parallel lapw0 at 三 10月 29 19:14:48 CST
2003
 --------
running lapw0 in single mode
822.010u 3.800s 13:55.90 98.7% 0+0k 0+0io 2406pf+0w
  lapw1  -p (19:28:44) starting parallel lapw1 at 三 10月 29 19:28:44 CST
2003
 ->  starting parallel LAPW1 jobs at 三 10月 29 19:28:44 CST 2003 

running LAPW1 in parallel mode (using .machines)
6 number_of_parallel_jobs
    c3(32)      c3(32)      c4(32)      c4(32)      c2(32)      c2(32)   
Summary of lapw1para:
  c3 k=32 user=0 wallclock=32
  c4 k=32 user=0 wallclock=32
  c2 k=32 user=0 wallclock=32
175.930u 279.330s 48:03:56.14 0.2% 0+0k 0+0io 22366465pf+0w
  lapw2 -p (19:32:41) running LAPW2 in parallel mode
     c3
     c3
     c4
     c4
     c2
     c2
  Summary of lapw2para:
  c3 user=0 wallclock=0
  c4 user=0 wallclock=0
  c2 user=0 wallclock=0
463.270u 3.570s 1:28:22.80 8.8% 0+0k 0+0io 29020pf+0w
  lcore (21:01:03) 1.790u 0.200s 0:04.28 46.4% 0+0k 0+0io 178pf+0w
  mixer (21:01:09) 21.940u 1.570s 0:28.09 83.6% 0+0k 0+0io 247pf+0w
:ENERGY convergence:  0 0 0
:CHARGE convergence:  0 0.001 0 

   cycle 2 (五 10月 31 21:01:38 CST 2003) 

   cycle 2 (19/19 to go)
  lapw0 -p (21:01:38)
starting parallel lapw0 at 五 10月 31 21:01:38 CST 2003
 --------
running lapw0 in single mode
819.280u 1.290s 13:45.70 99.3% 0+0k 0+0io 2522pf+0w
  lapw1  -p (21:15:24) starting parallel lapw1 at 五 10月 31 21:15:24 CST
2003
 ->  starting parallel LAPW1 jobs at 五 10月 31 21:15:24 CST 2003
running LAPW1 in parallel mode (using .machines)
6 number_of_parallel_jobs
    c3(32)      c3(32)      c4(32)      c4(32)      c2(32)      c2(32)   
Summary of lapw1para:
  c3 k=32 user=0 wallclock=32
  c4 k=32 user=0 wallclock=32
  c2 k=32 user=0 wallclock=32
191.310u 301.080s 45:45:32.04 0.2% 0+0k 0+0io 21963404pf+0w
  lapw2 -p (19:00:56) running LAPW2 in parallel mode
**  LAPW2 crashed!
1.620u 0.320s 1:20:49.69 0.0% 0+0k 0+0io 21233pf+0w 

  stop error

SHOW STDOUT
FORTRAN STOP  LAPW0 END
FORTRAN STOP  LAPW1 END 

real 2725m49.286s
user 2706m44.650s
sys 6m59.290s
FORTRAN STOP  LAPW1 END 

real092726m28.911s
user 2703m41.000s
sys 5m38.590s
FORTRAN STOP  LAPW1 END 

real 2868m56.009s
user 2839m55.520s
sys 7m5.360s
FORTRAN STOP  LAPW1 END 

real 2869m8.853s
user 2852m13.960s
sys 5m33.400s
FORTRAN STOP  LAPW1 END 

real 2883m40.191s
user 2859m4.590s
sys 5m45.970s
FORTRAN STOP  LAPW1 END 

real 2884m8.337s
user 2863m30.270s
sys 7m0.300s
FORTRAN STOP LAPW2 - FERMI; weighs written
FORTRAN STOP  LAPW2 END 

real 77m29.261s
user 74m17.980s
sys 1m32.470s
FORTRAN STOP  LAPW2 END 

real 77m41.464s
user 74m22.360s
sys 1m29.040s
FORTRAN STOP  LAPW2 END 

real 78m31.442s
user 75m39.420s
sys 1m36.280s
FORTRAN STOP  LAPW2 END 

real 78m53.943s
user 76m12.590s
sys 1m22.420s
FORTRAN STOP  LAPW2 END 

real 79m0.769s
user 76m34.110s
sys 1m20.660s
FORTRAN STOP  LAPW2 END 

real 80m22.742s
user 77m16.020s
sys 1m29.210s
FORTRAN STOP  SUMPARA END
FORTRAN STOP  SUMPARA END
FORTRAN STOP  CORE  END
FORTRAN STOP  MIX
ER END
in cycle 2    ETEST: 0   CTEST: 0
FORTRAN STOP  LAPW0 END
   full diagonalization forced
FORTRAN STOP  LAPW1 END 

real 2652m59.244s
user 2618m10.810s
sys 8m12.340s
FORTRAN STOP  LAPW1 END 

real 2666m22.835s
user 2648m27.370s
sys 4m58.030s
FORTRAN STOP  LAPW1 END 

real 2692m34.858s
user 2669m36.380s
sys 6m6.250s
FORTRAN STOP  LAPW1 END 

real 2721m6.973s
user 2693m20.250s
sys 6m32.880s
FORTRAN STOP  LAPW1 END 

real 2745m21.284s
user 2715m52.240s
sys 7m5.880s
FORTRAN STOP  LAPW1 END 

real 2745m41.973s
user 2720m46.660s
sys 6m48.260s
FORTRAN STOP LAPW2 - FERMI; weighs written 

Input/Output Error 154: Wrong length record 

  In Procedure: read_vec
       At Line: 95 

     Statement: Unformatted READ
          Unit: 10
  Connected To: ./sx509.vector_3
          Form: Unformatted
        Access: Sequential
Records Read   : 5419
Records Written: 0 

End of diagnostics 

real 1m10.277s
user 0m16.170s
sys 0m1.
520s
FORTRAN STOP  LAPW2 END 

real 75m46.218s
user 72m20.340s
sys 1m21.170s
FORTRAN STOP  LAPW2 END 

real 78m12.386s
user 74m21.870s
sys 1m29.240s
FORTRAN STOP  LAPW2 END 

real 79m17.077s
user 75m55.930s
sys 1m38.690s
FORTRAN STOP  LAPW2 END 

real 79m37.357s
user 75m39.350s
sys 1m30.320s
FORTRAN STOP  LAPW2 END 

real 80m46.623s
user 76m52.930s
sys 1m24.010s
cp: stat‘.in.tmp’failed: no this file or directory
rm: can`t delete‘.in.tmp’: no this file or directory
rm: can`t delete‘.in.tmp1’: no this file or directory. 

we have used 3 processors,every have 2 CPU,Mem:3098084k av ,Swap: 2040244k
av.
is it possible this big case? 

thank you for any suggestions! 

regards! 

           genghua
           hgeng at iccas.ac.cn 