Introduction¶

Asynchronous (async) execution defines the ability to run programs and functions in parallel and (possibly) independent of the main program flow. The proper use of async execution can greatly improve of performance of a program, and is only bounded by the physical number of CPUs and I/O limits.

This module & documentation provides the insights, concepts and tools within pyATS that supports asynchronous execution, allowing users to reap the full benefit of async without having to deal with its logistics & overhead.

those who don’t. There are 0b10 types of people in the world: those who understand asynchronous jokes and

Intentions¶

The intent of this module is not to reinvent asynchronous execution: Python’s built in library Threading and Multiprocessing already provides a great basis to do so, are quite powerful and user-friendly.

However, these libraries only provides a foundation to further build on. It is still up to the end user to write boilerplate code to handle and configure things like:

managing of log files

re-establishing lost connections, sessions

resource sharing & locking (such as telnet/ssh connections)

two-way inter-process communication and state synchronization

Therefore, the intent of this module is:

to strengthen pyATS’s support of asynchronous execution, automate some of the above items, and further streamlining the user experience.

However, keep in mind that not everything can be done within this module alone: in order to properly support asynchronous execution, each module/package has to be individually overhauled with parallelism in mind. Thus, this module documentation also serves as a central location where these concepts and changes are consolidated and shared in a overviewing fashion.