Distributed systems - Deepstash

🌍

Distributed systems

13 STASHED IDEAS

Personalised feed for everyone

Your Linkedin feed is based on your network, on your industry, content important to your company and colleagues as well.

It's a mix of algorithms and editors. 

  • Level of personalization and relevance coming from algorithms.
  • Editors burst the filter bubble. Give you things that you don't care about but are actually going to be very impactful to your professional lives.

Linkedin has editors focused on specific industries, like healthcare, finance, or engineering. 

3

STASHED IN:

8

Celery components
  • Client - Service that creates tasks
  • Task - Unit of work. When a task is submitted to the broker, it gets an ID and it's stored in a serialized format.
  • Worker - Daemon that spawns sub-processes to execute tasks (default number of sub-processes = number of CPU cores)
  • Broker - Message transport between clients and workers. Examples: Redis, RabbitMQ
  • Result Store - Database where task results are stored

STASHED IN:

4

The Realtime Platform

LinkedIn has built the Realtime Platform to distribute multiple types of data in real-time such as:

  • Likes, comments and viewer count for Live Videos
  • Typing indicators and Read receipts for Instant Messaging
  • Presence i.e. the green online indicators

Their goal is to increase user engagement by enabling dynamic instant experiences between users, such as: likes, comments, polls, discussions etc.

STASHED IN:

4