Goal: 

• As a developer, I need a means to assess in A/B fashion, the impact of removing the eventlet.monkey_patch(). As such, in order to facilitate testing, documentation updates, and any overt performance regression, I need to have a workload test which can be executed and thus measured on a basis to provide input.

Acceptance Criteria:

• A basic, re-producable test which can be leveraged to generate basic performance numbers. Some prior art exists for the creation of nodes, but measurement is focused on API response generation.

This may require additional insturmentation, or logging which could be then collated into results. Thread start and exit timing may be only one aspect to measure, but the idea is to get an idea of how drastically  (or not drastically at all) the performance shifts on lower level conductor process execution.