PROBLEM: To insert data in postgres SQL database and generating suitable response while managing burst requests

SOLUTION: The Idea is to queue requests(work) , use workers to process requests , dispatchers to assign work to workers,provide response using shared cache(to keep track of the request processed)

Key-Terms:

Collector:

    Gets request and pushes the required information to the queue(WorkQueue)

Work Queue:

    Use a buffered channel so the collector is not blocked

Dispatcher:

    Dispatcher distributes work among workers from WorkQueue

Workers:

    Responsible for storing data in database  and updating shared cache

Shared cache:

    Used to keep track of the request stored in the database ,once the response is generated the ID is removed from the cache

SETUP

    move to the /assignment directory
    $ go build -o queued *.go
    $ ./queued 

RESULTS :Load tested with Vegeta, capable of processing 20-24 concurrent requests/s with timeouts of 60s

IMPROVEMENTS:

1)Further improvements can be made by having more servers with load balancing

2)Instead of waiting for the response and showing timeout after a time interval ,we can also use an event triggered (event being a successful insertion in database) solution,where we can generate an SMS response to the user whenever the data gets inserted in the database.