further restrict IsolatedProcess by setting multiprocessing start met…

…hod (#181)

* further restrict IsolatedProcess by multiprocessing fork

* update docstyle

* update docstring

bluecellulab/simulation/parallel.py

@@ -1,6 +1,7 @@
 """Controlled simulations in parallel."""
 
 from __future__ import annotations
+import multiprocessing
 from multiprocessing.pool import Pool
 from bluecellulab.simulation.neuron_globals import NeuronGlobals
 
@@ -11,7 +12,8 @@ class IsolatedProcess(Pool):
     Use this when running isolated NEURON simulations. Running 2 NEURON
     simulations on a single process is to be avoided. Required global
     NEURON simulation parameters will automatically be passed to each
-    worker.
+    worker. `fork` mode of multiprocessing is used to pass any other
+    global NEURON parameters to each worker.
     """
 
     def __init__(self, processes: int | None = 1):
@@ -21,12 +23,14 @@ def __init__(self, processes: int | None = 1):
             processes: The number of processes to use for running the simulations.
                        If set to None, then the number returned by os.cpu_count() is used.
         """
+        ctx = multiprocessing.get_context("fork")
         neuron_global_params = NeuronGlobals.get_instance().export_params()
         super().__init__(
             processes=processes,
             initializer=self.init_worker,
             initargs=(neuron_global_params,),
             maxtasksperchild=1,
+            context=ctx,
         )
 
     @staticmethod