This is documentation for Mathematica 8, which was
based on an earlier version of the Wolfram Language.
View current documentation (Version 11.2)


launches all currently configured parallel subkernels.
launches n local subkernels on the current computer.
launches a subkernel with the given description.
launches several subkernels with the given descriptions.
  • Additional configurations for kernels on remote machines can be made available with the Wolfram Lightweight Grid and the Cluster Integration Package.
Launch the default kernels:
Close all running kernels and start two new ones:
Launch the default kernels:
Click for copyable input
Close all running kernels and start two new ones:
Click for copyable input
Launch the default kernels as described in $ConfiguredKernels:
Launch two more local kernels, in addition to those already running:
Close all kernels:
Launch copies of the previously running kernels:
Measure the speedup of the same calculation with different numbers of kernels used:
Distributed definitions and shared variables apply to running kernels and new ones:
Packages read with ParallelNeeds also apply to running and new kernels:
Close all running kernels and launch new ones:
The new kernels inherit previously distributed definitions and shared variables:
The new kernels also inherit packages read previously:
If any kernels are already running, LaunchKernels does not launch the default kernels:
Close all running kernels, then launch the default set of kernels:
Kernels may be automatically launched when first needed:
Manually close the kernels:
Now, no kernels are launched automatically:
If no kernels are available, most parallel calculations are done sequentially:
No kernels are launched by default on a single-core computer:
Specify the number of desired kernels to launch them anyway:
New in 7