Leetcode Submissions Simulation

Context

• Every day, millions of users across the globe use LeetCode for practicing DSA.
• for a curious Software Developer, questions might come like:
• How does LeetCode handle these submissions efficiently?
• Does a single server handles all the processing?
• Let's see how it's handled!
• This is not a complete but partial implementation of how LeetCode may handle submissions.

Technology Involved

Microservice Architecture

1. Currently, major web applications use the distributed servers architecture, also famously called Micro-service Architecture.

2. These servers are mainly utilized to:

   . Reduce the load of primary servers
   . Handle users based on geographic distribution 
   . Break down services to different servers to increase availability
   

Asynchronous Communications

• In asynchronous communications, the sender doesn't expect an immediate response from the other participants.
• Allowing Non-blocking communications, and enhances user experience by allowing them to perform other activities instead of waiting.

Queues & Pub-Subs

• Queues and pubsubs enable better handling of distributed and async communications across multiple servers.
• Redis is an open-source app, which enables features like: Database, caching and message brooking in memory(RAM).

Any major web-app these days utlizes these technologies to enhance metrics like: Availability and Scalability.

Project Architecture

• Primary Backend: Handles the requests sent from the user and pushes them into the redis queue or publishes them to the workers (Lpush).

• Worker: Handles the processing of the code by picking up available tasks in the queue or subscribing to the event and returns the result by publishing it to the primary backends*.

• This architecture of Primary backend - Worker reduces the load massively because processing of code can be quite intensive, and that to handling processing from multiple users can be computationally & time wise very expensive. That's why multiple worker servers only handle processing of submissions.

• This also increases availability, because if the primary backend were handling all the processing and it goes down, then the entire application goes down as well.

• Note: Here, generally pub-sub architecture is used to handle submission distribution & response management between the primary backends and the worker servers.