P3s
Revision as of 21:34, 4 January 2018 by MaximPotekhin (talk | contribs)
Overview
General notes on design and deployment (some of historical interest) can be found on a separate page: "P3s notes ". On that page, the basics of the pilot framework approach to p3s design and implementation are explained.
What is presented below is a brief "how-to" for p3s.
Clients and Servers
The p3s server is a Web service i.e. its API is based on HTTP requests. The user interacts with the system by means of clients which send HTTP requests to the server. This is the only way to change the state of the server and the objects it contains it its database.
The p3s server provides the following set of functionality:
- creates objects e.g. job records in its database, according to messages received from clients which may be actuated by the user or run as a part of an automated script
- changes the state of objects - once again, according to messages coming from the clients
- monitoring of the state of various objects such as jobs, pilots, units of data etc
p3s objects
Job
The "job" is p3s is a record in its database which contains all information necessary to create an actual process on a worker node.