##// END OF EJS Templates
mark 0.13.1.rc2
mark 0.13.1.rc2

File last commit:

r6455:15863dc1
r8563:8a2ea353
Show More
itermapresult.py
67 lines | 2.3 KiB | text/x-python | PythonLexer
MinRK
expand itermapresult docstring
r5217 """Example of iteration through AsyncMapResults, without waiting for all results
When you call view.map(func, sequence), you will receive a special AsyncMapResult
object. These objects are used to reconstruct the results of the split call.
One feature AsyncResults provide is that they are iterable *immediately*, so
you can iterate through the actual results as they complete.
This is useful if you submit a large number of tasks that may take some time,
but want to perform logic on elements in the result, or even abort subsequent
tasks in cases where you are searching for the first affirmative result.
By default, the results will match the ordering of the submitted sequence, but
if you call `map(...ordered=False)`, then results will be provided to the iterator
on a first come first serve basis.
MinRK
update parallel docs with some changes from scipy tutorial...
r5169
Authors
-------
* MinRK
"""
Thomas Kluyver
Update print syntax in parallel examples.
r6455 from __future__ import print_function
MinRK
update parallel docs with some changes from scipy tutorial...
r5169 import time
from IPython import parallel
# create client & view
rc = parallel.Client()
dv = rc[:]
v = rc.load_balanced_view()
# scatter 'id', so id=0,1,2 on engines 0,1,2
dv.scatter('id', rc.ids, flatten=True)
Thomas Kluyver
Update print syntax in parallel examples.
r6455 print("Engine IDs: ", dv['id'])
MinRK
update parallel docs with some changes from scipy tutorial...
r5169
# create a Reference to `id`. This will be a different value on each engine
ref = parallel.Reference('id')
Thomas Kluyver
Update print syntax in parallel examples.
r6455 print("sleeping for `id` seconds on each engine")
MinRK
update parallel docs with some changes from scipy tutorial...
r5169 tic = time.time()
ar = dv.apply(time.sleep, ref)
for i,r in enumerate(ar):
Thomas Kluyver
Update print syntax in parallel examples.
r6455 print("%i: %.3f"%(i, time.time()-tic))
MinRK
update parallel docs with some changes from scipy tutorial...
r5169
def sleep_here(t):
import time
time.sleep(t)
return id,t
# one call per task
Thomas Kluyver
Update print syntax in parallel examples.
r6455 print("running with one call per task")
MinRK
update parallel docs with some changes from scipy tutorial...
r5169 amr = v.map(sleep_here, [.01*t for t in range(100)])
tic = time.time()
for i,r in enumerate(amr):
Thomas Kluyver
Update print syntax in parallel examples.
r6455 print("task %i on engine %i: %.3f" % (i, r[0], time.time()-tic))
MinRK
update parallel docs with some changes from scipy tutorial...
r5169
Thomas Kluyver
Update print syntax in parallel examples.
r6455 print("running with four calls per task")
MinRK
update parallel docs with some changes from scipy tutorial...
r5169 # with chunksize, we can have four calls per task
amr = v.map(sleep_here, [.01*t for t in range(100)], chunksize=4)
tic = time.time()
for i,r in enumerate(amr):
Thomas Kluyver
Update print syntax in parallel examples.
r6455 print("task %i on engine %i: %.3f" % (i, r[0], time.time()-tic))
MinRK
update parallel docs with some changes from scipy tutorial...
r5169
Thomas Kluyver
Update print syntax in parallel examples.
r6455 print("running with two calls per task, with unordered results")
MinRK
update parallel docs with some changes from scipy tutorial...
r5169 # We can even iterate through faster results first, with ordered=False
amr = v.map(sleep_here, [.01*t for t in range(100,0,-1)], ordered=False, chunksize=2)
tic = time.time()
for i,r in enumerate(amr):
Thomas Kluyver
Update print syntax in parallel examples.
r6455 print("slept %.2fs on engine %i: %.3f" % (r[1], r[0], time.time()-tic))