thedevis / encrypted-timeseries

encrypted-timeseries

Problem Statement

Make a small backend application which can generate and emit an encrypted data stream over a socket, listens to incoming data stream on a socket, decrypts and decodes it, save to a time series db and then emit the saved data to a small frontend app. Backend services can be in any language (Node JS is preferred, but use Go or any other that you are comfortable with).

Emitter service

This service should generate a periodic data stream of encrypted messages where the number of encrypted strings can be anywhere between 49-499 so randomise this to good effect.

Each message should contain an object with 3 keys: name, origin, destination and a secret_key which is a hash of these 3 fields. randomize the values for name, origin, destination from a constant list provided in data.json file

exmaple

originalMessage = {
  name: 'Jack Reacher',
  origin: 'Bengaluru',
  destination: 'Mumbai'
}

add a secret_key by creating a sha-256 hash of the above object

sumCheckMessage = {
  name: 'Jack Reacher',
  origin: 'Bengaluru',
  destination: 'Mumbai',
  secret_key: "sha-256 ash of originalMessage",
}

encrypt this payload using encryption algorithm `aes-256-ctr` with a pass key

encryptedMessage = "encrypted message string"

the data stream is long string of | separated encrypted messages and can look like this

"e84742dedd1ddc924e5bfe9a5d912a1918e217f98e5578d04fd5c12426022240|4bbf088f4fc646d7a65b1f84172a59f665a09beb226368ff53d46a5edfd75dc6|3743c3ff07694a3e5540dfc14d57dcfdd6868439f9b5b83162be9162d8032999|26ccd3d082227c49907af7d3e4f19aec764f73d20b73ca4337df818b68cf6975|8d5c45f45be31d657dd58ae4e2c8222f61a779ad11fe36da7b00511ac2b5c01a|e97451a0c72d4202915f6c43b48bc4c0a500851e4c71b66b51b3a588e6522316|99624125591ebecb2c4e34695bf8d1e8a36b73087fd0c8e6c4fad087fa244d5c|b70ed78f5befa9c64ecd9ddcb64f18868ba86debf6b833ce440bcb772be3171c|a9bec91a127fb7b76a462fadeac5090b8dc753841f1fd54ac758f4cdb9af5fc0|2c345c51005cd0b0df92b089dba17e82e321725f539b1cdfceebd6eab69c336a"

The emitter service should connect to the listener service over sockets and periodically send out a new message stream every 10s

Listener Service

The Listener service will allow an emitter to connect to it via sockets. On receipt of the encrypted message stream, the listener should decrypt this string and retrieve the data in the payload. Validate the objects using the secret_key to ensure data integrity. If the data integrity of any object is compromised then discard that operation and move on to the next in the queue.

On successful object data integrity validation, add a timestamp to that object and save it to a mongoDB collection modelled for saving time-series data where each document should be corresponding to the minute in which it is received. e.g. for data received for a person between 14:00 to 14:01 all records are added in a single document as a timeseries. Design the schema is such a manner to allow for optimal performance for aggregation timeseries queries.

Frontend

All the valid data saved should be displayed in a real-time manner on a small frontend app along with the success rate for data transmission and decoding

How will the assignment be evaluated?

• Functionality: Is the app functional and performant?
• Code Quality: Code Readability and structuring.
• Exception Handling: No crashes and proper error messages logged.
• Git Commit practices: Manage your code in github and make frequent commits.
• Bonus consideration will be given if tests are also written.
• Bonus for Dockerising the apps

Submission

Share the github repository and hosted app link with a readme on how to run the code. If you have any other doubts, get in touch with us