Skip to content

Computation of Execution Plans to translate Planning Alternatives to the Software State

madcau edited this page Sep 4, 2017 · 2 revisions

Preliminary

iObserve uses the Mape-k approach for autonomic computing. Thus far, we implemented the Monitoring and the Analysis phases of this cycle. Beside the Planning, the other missing crucial aspect is the Execution phase. At the beginning of the execution, an model that describes the desired application state is given to the execution phase. This state is the result of the planning phase. Now the execution needs to establish the new state. For this certain actions, like (un-)deployment and migrations, have to be taken. The list of actions that should be taken to create the new state is called an Execution Plan. It 'translates' the planned state of the system to the real state. In this task you will design and implement the generation of the execution plan and also implement all possible actions needed to execute this plan.

Related Repositories

https://github.com/research-iobserve/iobserve-analysis
https://github.com/research-iobserve/iobserve-ui-deployment

Documentation

Write me :)