Job Handling

Overview

When we run a simulation (which we will refer as a job), we need to make a request to the simulation server (a python server) that runs the simulation in the background, the simulation takes some time to run, so we do not wait for the simulation to finish, instead we tell the user that the simulation is running.

Once a job is invoked, the simulation server will start running the simulation, we require that the simulation server outputs the files:

• <jobId>.json: a file that contains the result of the simulation

The front end will poll the simulation server for the progress of the simulation, where the simulation server will return the progress.json file, the front end will then update the progress bar accordingly. If the simulation is finished, the front end will then request the result.json file, and display the result to the user.

Simulation files

A job is tagged with a jobId which is used to identify the job. Then the JSON files will be saved under the directory ./results/.

├── <jobId>.json

Progress file

The schema for the progress file is as follows:

{
  "progress": 0.5,
  "id": "<jobId>"
  "created": 1620000000, // unix timestamp
  "num_reps": "1", // number of repititions
  "results": [
    {...}, // Scenario 1
    {...}, // Scenario 2
    {...}, // Scenario 3
    ...
  ]
}

The JSON schema for scenarios can be found in the JSON schema page