about replica exchange example in lib of 2.10 NAMD

From: jc guo (guojc123_at_gmail.com)
Date: Fri Oct 16 2015 - 00:11:18 CDT


I saw some were asking about this,

but I did not get clear answer from above, is the replica exchange not work
for single node? only works on the cluster? since I do not know how to find
my nodelist in our workstation, so could you help me.

I download source code of NAMD 2.10, and follow the compile step in
note.txt, such as,

  tar xzf NAMD_2.10_Source.tar.gz
  cd NAMD_2.10_Source
  tar xf charm-6.6.1.tar

Build and test the Charm++/Converse library (MPI version):
  cd charm-6.6.1
  env MPICXX=mpicxx ./build charm++ mpi-linux-x86_64 --with-production
  cd mpi-linux-x86_64/tests/charm++/megatest
  make pgm
  mpirun -n 4 ./pgm (run as any other MPI program on your cluster)
  cd ../../../../..

Download and install TCL and FFTW libraries:
  (cd to NAMD_2.10_Source if you're not already there)
  tar xzf fftw-linux-x86_64.tar.gz
  mv linux-x86_64 fftw
  tar xzf tcl8.5.9-linux-x86_64.tar.gz
  tar xzf tcl8.5.9-linux-x86_64-threaded.tar.gz
  mv tcl8.5.9-linux-x86_64 tcl
  mv tcl8.5.9-linux-x86_64-threaded tcl-threaded

Optionally edit various configuration files:
  (not needed if charm-6.6.1, fftw, and tcl are in NAMD_2.10_Source)
already in same directory.

Set up build directory and compile:
  MPI version: ./config Linux-x86_64-g++ --charm-arch mpi-linux-x86_64
  cd Linux-x86_64-g++
  make (or gmake -j4, which should run faster)

Quick tests using one and two processes (network version):
  (this is a 66-atom simulation so don't expect any speedup)
  ./namd2 src/alanin
  ./charmrun +p2 ./namd2
  ./charmrun +p2 ./namd2 src/alanin

Longer test using four processes:
  wget http://www.ks.uiuc.edu/Research/namd/utilities/apoa1.tar.gz
  tar xzf apoa1.tar.gz
  ./charmrun +p4 ./namd2 apoa1/apoa1.namd

above two test case works, but when I ran replica module in NAMD, follows
such steps,

          cd example (in replica directory)
          mkdir output
          (cd output; mkdir 0 1 2 3 4 5 6 7)
          mpirun namd2 +replicas 8 job0.conf +stdout output/%d/job0.%d.log
error happens, likes,

Converse/Charm++ Commit ID:
Warning> Randomization of stack pointer is turned on in kernel, thread
migration may not work! Run 'echo 0 >
/proc/sys/kernel/randomize_va_space' as root to disable it, or try run
with '+isomalloc_sync'.
CharmLB> Load balancer assumes all CPUs are same.
Charm++> Running on 1 unique compute nodes (8-way SMP).
Charm++> cpu topology info is gathered in 0.001 seconds.
Info: NAMD 2.10 for Linux-x86_64-MPI
Info: Please visit http://www.ks.uiuc.edu/Research/namd/
Info: for updates, documentation, and support information.
Info: Please cite Phillips et al., J. Comp. Chem. 26:1781-1802 (2005)
Info: in all publications reporting results obtained with NAMD.
Info: Based on Charm++/Converse 60601 for mpi-linux-x86_64
Info: Built Mon Sep 28 02:01:43 SGT 2015 by tanmingchen on tanmingchen
Info: 1 NAMD 2.10 Linux-x86_64-MPI 1 tanmingchen tanmingchen
Info: Running on 1 processors, 1 nodes, 1 physical nodes.
Info: CPU topology information available.
Info: Charm++/Converse parallel runtime startup completed at 0.0263531 s
Info: 83.3789 MB of memory in use based on /proc/self/stat
Info: Configuration file is ../lib/replica/example/job0.conf
Info: Changed directory to ../lib/replica/example
replica_neighbors proc passes internal consistency check
TCL: Reduction callback proc set to save_callback
TCL: can't read "temperature": no such variable
FATAL ERROR: can't read "temperature": no such variable
    while executing
"temperature $temperature"
    invoked from within
"if {$i_run} { #restart
  bincoordinates $restart_root.$replica_id.coor
  binvelocities $restart_root.$replica_id.vel
  extendedSystem $restart_root.$r..."
    (file "~/gg/namd/NAMD_2.10_Source/lib/replica/umbrella.namd" line 90)
    invoked from within
"source ~/gg/namd/NAMD_2.10_Source/lib/replica/umbrella.namd"
    invoked from within
"if { ! [catch numPes] } { source
    (file "job0.conf" line 5)
[0] Stack Traceback:
  [0:0] CmiAbortHelper+0x63 [0xc56ca3]
  [0:1] _Z8NAMD_diePKc+0x84 [0x5e5a94]
  [0:2] _Z18after_backend_initiPPc+0x212 [0x5ea7d2]
  [0:3] main+0x1a [0x57f34a]
  [0:4] __libc_start_main+0xf5 [0x7f7e901dfec5]
  [0:5] [0x58217c]

could you help me? Is replica only can work on cluster, not support my
workstation's information looks like?

 grep 'core id' /proc/cpuinfo | sort -u | wc -l
grep 'processor' /proc/cpuinfo | sort -u | wc -l

Sorry to disturb you, hope I could get some help from any of you.

Best regards,
JC Guo

This archive was generated by hypermail 2.1.6 : Tue Dec 27 2016 - 23:21:24 CST