p3s stands for the ProtoDUNE Prompt Processing System.
It was initially developed as a lightweight and easy-to-deploy tool for managing prompt processing workflows in the protoDUNE experiment at CERN. However, p3s does not contain any special logic or dependency related specifically to protoDUNE. It is content-agnostic and can be utilized for many other purposes.
The design of p3s was inspired by pilot-based frameworks such as PanDA and Dirac, but p3s is a clean sheet development and prioritizes simplicity and component reuse over most other factors.
Important subdirectories:
- documents: documentation
- promptproc: the Web application server code
- clients: the p3s client code
- inputs: useful examples of actual job description both for local testing and for the real operating environment
- configuration: templates of the p3s environment, which will likely be site-dependent
Misc:
- display: the Web application which is a companion of p3s, used to display results of calculations specific to protoDUNE data quality management
Please consult the documents subdirectory for documentation concerning
- system overview
- installation of p3s and its software dependencies
- p3s client interface (important for operators and shifters)
- advanced part of the design such as workflow functionality etc
The documents are best viewed and/or printed out in the PDF format. These files are kept in documents/pdf folder.
In addition, don't forget that most p3s clients are fairly well self-documented via the use of "-h" option which prints out useful synopsis for each command.