Given a limited set of resources and users requesting access to them, assign the resources to users in such a way that a resource is assigned to at most a user at a time.

• the resources are loaded in a sqlite database from a csv file
• to prevent different users from modifying the same file at the same time, every operation is encapsulated in a connection to the database
• log every operation in a log.txt file
• to change the locations of the output files, edit src/constants.py

Initialize the db and the manager: ./clean.sh && ./init.sh

Display the database content: ./print_db.py

Make requests: ./request_from_input.py

Free a resource: ./free_from_input.py

The file input/requests.csv contains a scenario of ~40 requests which can be used to run a simulation

• cd src/
• ./clean.sh && ./init.sh
• ./simulation.py

• implement a priority system
  • for example, assign low priority to automated task and high to people
  • it only affects the queueing system and it can be easily implemented by keeping separate queues for every priority level
• assign to each node a set of label and let the user refer to them
• consider some cost assigned to each node and find the minimal cost assignment
  • an example of cost might be the use-history of each node
  • ideally, consider an explicit cost function to minimize (taking possibly everything into account: priorities, waiting times, ...)
• the current implementation (and especially the queueing system) is limited by the fact that there isn't a server running in the background

Q: Should I use it in a real scenario?
A: Please don't The system works well enough, but it probably needs further testing and development.