Skip to content

Latest commit

 

History

History
24 lines (20 loc) · 1.76 KB

README.md

File metadata and controls

24 lines (20 loc) · 1.76 KB

Django Twitter Backend Service

Introduction

A backend social networking service based on Python and Django, aiming for low latency and high throughput. It provides APIs for building friendships, posting messages and photos, news feeds, giving comments and likes, pushing notifications, etc.

  • Built low-latency SNS backend using Python, Django & RESTful API.
  • Implemented newsfeeds using push model, and implemented endless pagination for mobile-friendly user experience.
  • Reduced queries for read-heavy tables by using Redis and Memcached.
  • Reduced database queries by using denormalization to store numbers of comments and likes.
  • Improved database query performance of friendships and newsfeeds by adopting HBase to support range queries.
  • Reduced response time by leveraging Redis as message queue to deliver asynchronized tasks.
  • Committed over 10000 lines of code changes, spent over 3 months to finish.

Technologies

Technology Description Website
Django Python-based MVC framework https://www.djangoproject.com/
DjangoRESTframework REST framework https://www.django-rest-framework.org/
MySQL Sql DB https://www.mysql.com/
HBase NoSql DB https://hbase.apache.org/
Memcached Distributed in-memory cache https://memcached.org/
Redis Distributed in-memory cache https://redis.io/
Amazon S3 Object storage service https://www.amazon.com/aws/s3