Initial Design of Quora

Initial design

This initial design separates request handling, persistent content, cached data, and ranking work. These are choices for the proposed service, not a verified inventory of Quora’s current infrastructure:

• Web and application servers: Web servers accept incoming HTTP requests and forward them to application servers for processing. They run manager processes that delegate work to worker processes on application servers through a routing layer. This routing layer manages task queues, where managers enqueue tasks and workers dequeue them for execution. Application servers may also maintain in-memory priority queues to differentiate request classes. The following diagram shows an abstract view of this architecture: