#!/usr/bin/env python
#
# Author: Mike McKerns (mmckerns @caltech and @uqfoundation)
# Copyright (c) 1997-2016 California Institute of Technology.
# Copyright (c) 2016-2018 The Uncertainty Quantification Foundation.
# License: 3-clause BSD. The full license text is available at:
# - https://github.com/uqfoundation/pathos/blob/master/LICENSE
from pathos.parallel import stats
from pathos.parallel import ParallelPool as Pool
pool = Pool()
def host(id):
import socket
import time
time.sleep(1.0)
return "Rank: %d -- %s" % (id, socket.gethostname())
print("Evaluate 10 items on 2 cpus") #FIXME: reset lport below
pool.ncpus = 2
pool.servers = ('localhost:5653', )
res5 = pool.map(host, range(10))
print(pool)
print('\n'.join(res5))
print(stats())
print('')
# end of file
#
# Author: Mike McKerns (mmckerns @caltech and @uqfoundation)
# Copyright (c) 1997-2016 California Institute of Technology.
# License: 3-clause BSD. The full license text is available at:
# - http://trac.mystic.cacr.caltech.edu/project/pathos/browser/pathos/LICENSE
from pathos.parallel import stats
from pathos.parallel import ParallelPool as Pool
pool = Pool()
def host(id):
import socket
return "Rank: %d -- %s" % (id, socket.gethostname())
print "Evaluate 10 items on 1 cpu"
pool.ncpus = 1
res3 = pool.map(host, range(10))
print pool
print '\n'.join(res3)
print stats()
print "Evaluate 10 items on 2 cpus"
pool.ncpus = 2
res5 = pool.map(host, range(10))
print pool
print '\n'.join(res5)
print stats()
# end of file
# Author: Mike McKerns (mmckerns @caltech and @uqfoundation)
# Copyright (c) 1997-2016 California Institute of Technology.
# License: 3-clause BSD. The full license text is available at:
# - http://trac.mystic.cacr.caltech.edu/project/pathos/browser/pathos/LICENSE
from pathos.parallel import stats
from pathos.parallel import ParallelPool as Pool
pool = Pool()
def host(id):
import socket
return "Rank: %d -- %s" % (id, socket.gethostname())
print "Evaluate 10 items on 1 cpu"
pool.ncpus = 1
res3 = pool.map(host, range(10))
print pool
print '\n'.join(res3)
print stats()
print "Evaluate 10 items on 2 cpus"
pool.ncpus = 2
res5 = pool.map(host, range(10))
print pool
print '\n'.join(res5)
print stats()
# end of file
# Author: Mike McKerns (mmckerns @caltech and @uqfoundation)
# Copyright (c) 1997-2016 California Institute of Technology.
# Copyright (c) 2016-2019 The Uncertainty Quantification Foundation.
# License: 3-clause BSD. The full license text is available at:
# - https://github.com/uqfoundation/pathos/blob/master/LICENSE
from pathos.parallel import stats
from pathos.parallel import ParallelPool as Pool
pool = Pool()
def host(id):
import socket
return "Rank: %d -- %s" % (id, socket.gethostname())
print("Evaluate 10 items on 1 cpu")
pool.ncpus = 1
res3 = pool.map(host, range(10))
print(pool)
print('\n'.join(res3))
print(stats())
print("Evaluate 10 items on 2 cpus")
pool.ncpus = 2
res5 = pool.map(host, range(10))
print(pool)
print('\n'.join(res5))
print(stats())
# end of file
def evaluate_distributed(self,
func,
*args,
nprocs=1,
servers=None,
verbose=False,
**kwargs):
"""Distribute model evaluation across a cluster.
Usage Conditions:
* The provided function needs to accept a numpy array of inputs as
its first parameter
* The provided function must return a numpy array of results
Parameters
----------
func : function,
Model, or function that wraps a model, to be run in parallel
nprocs : int,
Number of processors to use for each node. Defaults to 1.
servers : list[str] or None,
IP addresses or alias for each server/node to use.
verbose : bool,
Display job execution statistics. Defaults to False.
*args : list,
Additional arguments to be passed to `func`
**kwargs : dict,
Additional keyword arguments passed to `func`
Returns
----------
self : ProblemSpec object
"""
if verbose:
from pathos.parallel import stats
warnings.warn(
"This is an untested experimental feature and may not work.")
workers = pp_Pool(nprocs, servers=servers)
# Split into even chunks
chunks = np.array_split(self._samples,
int(nprocs) * len(servers),
axis=0)
tmp_f = self._wrap_func(func)
res = list(workers.map(tmp_f, chunks))
self._results = self._collect_results(res)
if verbose:
print(stats(), '\n')
workers.clear()
return self