Getting Started#


It is recommend that you install hyper-shell inside a virtual environment. See the installation guide for more options and additional notes and recommendations.

pipx install hyper-shell
brew tap glentner/tap
brew install hyper-shell

Basic Usage#

Complete details on all execution modes, parallelism, and options are available in the command-line documentation. Complete, specific examples are also available in the tutorials section.

In most cases, using the cluster subcommand is best. If you have some file,, that lists shell commands that you might otherwise execute alone (which would run each line in serial), pass that file to hyper-shell to process those commands in parallel.

Basic Usage

hyper-shell cluster

To specify the number of tasks to execute simultaneously, use --num-cores (or -N for short).

Parallel Workers

hyper-shell cluster -N16

Assuming the individual commands run on a single-core (they themselves are not parallel applications), you should use the same number as the number of physical cores on your system.

Some commands may fail for whatever reason. To track which input commands had a non-zero exit status, specify the --failed (or -f for short) option. This output file will contain lines from the input file that failed.

Track Failed Tasks

hyper-shell cluster -N16 -f tasks.failed