xgboost/tracker/rabit_sge.py
tqchen 75bf97b575 Squashed 'subtree/rabit/' changes from 091634b..59e63bc
59e63bc minor
6233050 ok
14477f9 add namenode
75a6d34 add libhdfs opts
e3c76bf minmum fix
8b3c435 chg
2035799 test code
7751b2b add debug
7690313 ok
bd346b4 ok
faba1dc add testload
6f7783e add testload
e5f0340 ok
3ed9ec8 chg
e552ac4 ask for more ram in am
b2505e3 only stop nm when sucess
bc696c9 add queue info
f3e867e add option queue
5dc843c refactor fileio
cd9c81b quick fix
1e23af2 add virtual destructor to iseekstream
f165ffb fix hdfs
8cc6508 allow demo to pass in env
fad4d69 ok
0fd6197 fix more
7423837 fix more
d25de54 add temporal solution, run_yarn_prog.py
e5a9e31 final attempt
ed3bee8 add command back
0774000 add hdfs to resource
9b66e7e fix hadoop
6812f14 ok
08e1c16 change hadoop prefix back to hadoop home
d6b6828 Update build.sh
146e069 bugfix: logical boundary for ring buffer
19cb685 ok
4cf3c13 Merge branch 'master' of ssh://github.com/tqchen/rabit
20daddb add tracker
c57dad8 add ringbased passing and batch schedule
295d8a1 update
994cb02 add sge
014c866 OK

git-subtree-dir: subtree/rabit
git-subtree-split: 59e63bc1354c9ff516d72d9a6468f6c431627202
2015-03-21 00:44:31 -07:00

70 lines
3.0 KiB
Python
Executable File

#!/usr/bin/env python
"""
Submit rabit jobs to Sun Grid Engine
"""
import argparse
import sys
import os
import subprocess
import rabit_tracker as tracker
parser = argparse.ArgumentParser(description='Rabit script to submit rabit job using MPI')
parser.add_argument('-n', '--nworker', required=True, type=int,
help = 'number of worker proccess to be launched')
parser.add_argument('-q', '--queue', default='default', type=str,
help = 'the queue we want to submit the job to')
parser.add_argument('-hip', '--host_ip', default='auto', type=str,
help = 'host IP address if cannot be automatically guessed, specify the IP of submission machine')
parser.add_argument('--vcores', default = 1, type=int,
help = 'number of vcpores to request in each mapper, set it if each rabit job is multi-threaded')
parser.add_argument('--jobname', default='auto', help = 'customize jobname in tracker')
parser.add_argument('--logdir', default='auto', help = 'customize the directory to place the logs')
parser.add_argument('-v', '--verbose', default=0, choices=[0, 1], type=int,
help = 'print more messages into the console')
parser.add_argument('command', nargs='+',
help = 'command for rabit program')
args = parser.parse_args()
if args.jobname == 'auto':
args.jobname = ('rabit%d.' % args.nworker) + args.command[0].split('/')[-1];
if args.logdir == 'auto':
args.logdir = args.jobname + '.log'
if os.path.exists(args.logdir):
if not os.path.isdir(args.logdir):
raise RuntimeError('specified logdir %s is a file instead of directory' % args.logdir)
else:
os.mkdir(args.logdir)
runscript = '%s/runrabit.sh' % args.logdir
fo = open(runscript, 'w')
fo.write('\"$@\"\n')
fo.close()
#
# submission script using MPI
#
def sge_submit(nslave, worker_args, worker_envs):
"""
customized submit script, that submit nslave jobs, each must contain args as parameter
note this can be a lambda function containing additional parameters in input
Parameters
nslave number of slave process to start up
args arguments to launch each job
this usually includes the parameters of master_uri and parameters passed into submit
"""
env_arg = ','.join(['%s=\"%s\"' % (k, str(v)) for k, v in worker_envs.items()])
cmd = 'qsub -cwd -t 1-%d -S /bin/bash' % nslave
if args.queue != 'default':
cmd += '-q %s' % args.queue
cmd += ' -N %s ' % args.jobname
cmd += ' -e %s -o %s' % (args.logdir, args.logdir)
cmd += ' -pe orte %d' % (args.vcores)
cmd += ' -v %s,PATH=${PATH}:.' % env_arg
cmd += ' %s %s' % (runscript, ' '.join(args.command + worker_args))
print cmd
subprocess.check_call(cmd, shell = True)
print 'Waiting for the jobs to get up...'
# call submit, with nslave, the commands to run each job and submit function
tracker.submit(args.nworker, [], fun_submit = sge_submit, verbose = args.verbose)