| 5 | 1/1 | 返回列表 |
| 查看: 3108 | 回復: 6 | |||
| 當前只顯示滿足指定條件的回帖,點擊這里查看本話題的所有回帖 | |||
[交流]
GAMESS最新安裝通過
|
|||
|
GAMESS最新安裝通過--不知有沒有遺漏,測試結果有些問題,請高手指點,謝謝 這些怎么沒有加入 測試結果 測試結果-- DDI Process 0: error code 911 ,不知何故 |
» 搶金幣啦!回帖就可以得到:
+1/483
+1/84
+1/71
+1/58
+2/46
+1/40
+2/22
+1/21
+1/17
+2/16
+1/14
+1/13
+1/9
+1/9
+1/8
+1/7
+1/6
+1/3
+1/2
+1/1
|
編譯程序時用的MPICH2,rungms里面的內(nèi)容修改搞的不是很明白,請高手指點一下?纯茨男┑胤叫枰薷模瑀ungms里默認的是Intel MPI rungms: Here we use two constant node names, compute-0-0 and compute-0-1, each of which is assumed to be SMP (ours are 8-ways): Each user must set up a file named ~/.mpd.conf containing a single line: "secretword=GiantsOverDodgers" which is set to user-only access permissions "chmod 600 ~/.mpd.conf". The secret word shouldn't be a login password, but can be anything you like: "secretword=VikingsOverPackers" is just as good. if ($TARGET == mpi) then # # Run outside of the batch schedular Sun Grid Engine (SGE) # by faking SGE's host assignment file: $TMPDIR/machines. # This script can be executed interactively on the first # compute node mentioned in this fake 'machines' file. set TMPDIR=$SCR # perhaps SGE would assign us two node names... echo "compute-0-1" > $TMPDIR/machines echo "compute-0-2" >> $TMPDIR/machines # or if you want to use these four nodes... #--echo "compute-0-0" > $TMPDIR/machines #--echo "compute-0-1" >> $TMPDIR/machines #--echo "compute-0-2" >> $TMPDIR/machines #--echo "compute-0-3" >> $TMPDIR/machines # # besides the usual three arguments to 'rungms' (see top), # we'll pass in a "processers per node" value. This could # be a value from 1 to 8 on our 8-way nodes. set PPN=$4 # # Allow for compute process and data servers (one pair per core) # @ NPROCS = $NCPUS + $NCPUS # # MPICH2 kick-off is guided by two disk files (A and B). # # A. build HOSTFILE, saying which nodes will be in our MPI ring # setenv HOSTFILE $SCR/$JOB.nodes.mpd if (-e $HOSTFILE) rm $HOSTFILE touch $HOSTFILE # if ($NCPUS == 1) then # Serial run must be on this node itself! echo `hostname` >> $HOSTFILE set NNODES=1 else # Parallel run gets node names from SGE's assigned list, # which is given to us in a disk file $TMPDIR/machines. uniq $TMPDIR/machines $HOSTFILE set NNODES=`wc -l $HOSTFILE` set NNODES=$NNODES[1] endif # uncomment these if you are still setting up... #--echo '------------' #--echo HOSTFILE $HOSTFILE contains #--cat $HOSTFILE #--echo '------------' # # B. the next file forces explicit "which process on what node" rules. # setenv PROCFILE $SCR/$JOB.processes.mpd if (-e $PROCFILE) rm $PROCFILE touch $PROCFILE # if ($NCPUS == 1) then @ NPROCS = 2 echo "-n $NPROCS -host `hostname` /home/mike/gamess/gamess.$VERNO.x" >> $PROCFILE else @ NPROCS = $NCPUS + $NCPUS if ($PPN == 0) then # when our SGE is just asked to assign so many cores from one # node, PPN=0, we are launching compute processes and data # servers within our own node...simple. echo "-n $NPROCS -host `hostname` /home/mike/gamess/gamess.$VERNO.x" >> $PROCFILE else # when our SGE is asked to reserve entire nodes, 1<=PPN<=8, # the $TMPDIR/machines contains the assigned node names # once and only once. We want PPN compute processes on # each node, and of course, PPN data servers on each. # Although DDI itself can assign c.p. and d.s. to the # hosts in any order, the GDDI logic below wants to have # all c.p. names before any d.s. names in the $HOSTFILE. # # thus, lay down a list of c.p. @ PPN2 = $PPN + $PPN @ n=1 while ($n <= $NNODES) set host=`sed -n -e "$n p" $HOSTFILE` set host=$host[1] echo "-n $PPN2 -host $host /home/mike/gamess/gamess.$VERNO.x" >> $PROCFILE @ n++ end endif endif # uncomment these if you are still setting up... #--echo PROCFILE $PROCFILE contains #--cat $PROCFILE #--echo '------------' # echo "MPICH2 will be running GAMESS on $NNODES nodes." echo "The binary to be kicked off by 'mpiexec' is gamess.$VERNO.x" echo "MPICH2 will run $NCPUS compute processes and $NCPUS data servers." if ($PPN > 0) echo "MPICH2 will be running $PPN of each process per node." # # Next sets up MKL usage setenv LD_LIBRARY_PATH /opt/intel/mkl/10.0.3.020/lib/em64t # force old MKL versions (version 9 and older) to run single threaded setenv MKL_SERIAL YES # setenv LD_LIBRARY_PATH /opt/mpich2/gnu/lib:$LD_LIBRARY_PATH set path=(/opt/mpich2/gnu/bin $path) # echo The scratch disk space on each node is $SCR chdir $SCR # # Now, at last, we can actually launch the processes, in 3 steps. # a) bring up a 'ring' of MPI demons # set echo mpdboot --rsh=ssh -n $NNODES -f $HOSTFILE # # b) kick off the compute processes and the data servers # mpiexec -configfile $PROCFILE < /dev/null # # c) shut down the 'ring' of MPI demons # mpdallexit unset echo # # HOSTFILE is passed to the file erasing step below rm -f $PROCFILE endif |
| 最具人氣熱帖推薦 [查看全部] | 作者 | 回/看 | 最后發(fā)表 | |
|---|---|---|---|---|
|
[考研] 362求調(diào)劑 +13 | 西南交材料專碩3 2026-03-31 | 13/650 |
|
|---|---|---|---|---|
|
[考研] 301求調(diào)劑 +10 | A_JiXing 2026-04-01 | 10/500 |
|
|
[考研] 08工科275分求調(diào)劑 +12 | AaAa7420 2026-03-31 | 12/600 |
|
|
[考研] 一志愿211,335分,0856,求調(diào)劑院校和導師 +11 | 傾____蕭 2026-03-27 | 12/600 |
|
|
[考研] 材料專碩306英一數(shù)二 +7 | z1z2z3879 2026-03-31 | 7/350 |
|
|
[考研] 296求調(diào)劑 +3 | 汪!?! 2026-03-31 | 6/300 |
|
|
[考研] 0710生物學求調(diào)劑! +6 | 敘述文 2026-03-31 | 6/300 |
|
|
[考研] 309分085801求調(diào)劑 +7 | 學員Gtwj7W 2026-03-31 | 7/350 |
|
|
[考研] 生物學308分求調(diào)劑(一志愿華東師大) +3 | 相信必會光芒萬?/a> 2026-03-31 | 3/150 |
|
|
[考研] 一志愿華東理工大學,080500學碩,317分,求調(diào)劑 +9 | s1145 2026-03-31 | 9/450 |
|
|
[碩博家園] 求調(diào)劑 有機化學考研356分 +11 | Nadiums 2026-03-25 | 12/600 |
|
|
[考研] 086000調(diào)劑 +5 | 7901117076 2026-03-26 | 5/250 |
|
|
[考研] 一志愿南昌大學324求調(diào)劑 +6 | hanamiko 2026-03-29 | 6/300 |
|
|
[考研] 0703化學321分求調(diào)劑 +10 | 三dd. 2026-03-30 | 11/550 |
|
|
[考研] 297求調(diào)劑 +17 | 田洪有 2026-03-26 | 18/900 |
|
|
[考研] 070305高分子化學與物理 304分求調(diào)劑 +12 | c297914 2026-03-28 | 12/600 |
|
|
[考研] 356求調(diào)劑 +4 | gysy?s?a 2026-03-28 | 4/200 |
|
|
[考研] 265求調(diào)劑11408 +3 | 劉小鹿lu 2026-03-27 | 3/150 |
|
|
[考研] 0856調(diào)劑 +5 | 求求讓我有書讀?/a> 2026-03-26 | 6/300 |
|
|
[考研] 298調(diào)劑 +3 | jiyingjie123 2026-03-27 | 3/150 |
|