Skip to content

Latest commit

 

History

History
55 lines (37 loc) · 3.19 KB

Problem Statement.md

File metadata and controls

55 lines (37 loc) · 3.19 KB

The public transport authority of Snowdonia town (fictitious) has mandated that all transport modes in the town be equipped with GPS emitters, radiating a vehicle's position every 20 seconds. The mayor has come to you to devise a solution for collecting this information for appropriate use by the city council to make intelligent future decisions about their public transport policy.

You are tasked with constructing a data collection endpoint to efficiently receive all GPS emissions from the city's fleet of:

  • buses
  • taxis
  • trams
  • trains

Every emission from the new devices will always contain the following pieces of data:

  • unique identifier for the emitting vehicle
  • vehicle type
  • position latitude
  • position longitude
  • timestamp of recorded position
  • current heading
    • Number between 0 and 359: 0 is true North. 180 is true South. 90 is true East. 270 is true West

Additional Business Requirements

  • If a vehicle has exited the "city boundaries", disregard any emissions from that vehicle.

    • City Boundary is currently assumed to be a 50km radius around an the town centre, an arbitrary point you select.
    • Just ensure your town centre is not 0 lat, 0 lng

  • All position data received must be stored somewhere for future analysis and consumption by the city council.

Testing Requirements

Snowdonia town has ~1000 vehicles in active operation at any given moment. Each vehicle could theoretically emit their position at the exact same moment. The solution must demonstrate evidence of 1000 concurrent emissions (at maximum) being handled by your developed endpoint without an error.

We will accept any form of evidence of successful testing. Some examples:

  • clearly-readable performance benchmarking
  • a video screen capture of your 1000 vehicle clients being received
  • your creative choice

We would like to see whatever testing mechanism you constructed to radiate GPS positions for your suite of 1000 test vehicles. Please ensure it is committed along with the endpoint code, with clear instructions for us to run your test client (bulk emitter).

Technical Assumptions

  • The city will implement whatever data transport mechanism & protocol you recommend on the GPS emitters, as long as:
    • The data payload contains the GPS data elements listed above
    • The data payloads are all directed to the endpoint you construct
  • You are not limited in the endpoint architecture, design, protocol, system or mechanism you wish to implement.
  • The solution is expected to be implemented using one of the following languages: Ruby, Python or Javascript.
  • Feel free to pre-seed your ecosystem with arbitrary vehicles and unique identifiers. Mechanism for registering vehicles in your solution are not part of the challenge.
  • Unique identifiers conform to the UUID identifier standard (https://en.wikipedia.org/wiki/Universally_unique_identifier)

Optional

A containerized solution is not a requirement, but definitely preferred for us to easily deploy your solution and test. Therefore, adding a Dockerfile to your solution is optional, but recommended. Ideally, your API would be deployed to a public location (heroku, AWS, your choice) for us to test against, with your 1000 client concurrent test suite.