Facebook墙的数据库结构 [英] Facebook wall's database structure

查看:197
本文介绍了Facebook墙的数据库结构的处理方法,对大家解决问题具有一定的参考价值,需要的朋友们下面随着小编来一起学习吧!

问题描述

我需要创建像Facebook的壁挂系统(用户可以发布消息,视频,图像,事件等)。有人创造了类似的东西吗?你如何建议我准备数据库?

I need to create a wall system just like Facebook's (users can post messages, videos, images, events, etc...). Has anyone created something similar? How do you recommend me to prepare the database?

推荐答案

更新后

Facebook已经介绍了他们对数据的使用情况。每个人都假设他们是管理大数据的先驱,这是现在的流行语。但实际上,他们的演讲题目是很多很多小数据。也就是说,他们基本上写了很多关于阅读或写个人记录的查询。这是一个简单的过滤,但是它的SQL数据库是一个巨大的键/值存储。这种架构的一个方面是,它使得它们能够直接与memcached中的副本交换SQL数据。

Facebook has given presentations talking about their use of data. Everyone assumes they are pioneers in the management of Big Data which is a buzzword these days. But in fact, their presentation was titled Lots and Lots of Small Data. That is, they basically write lots of queries for reading or writing individual records. This is an oversimplification, but it's like their SQL database is an enormous key/value store. One aspect of this architecture is that it makes it straightforward for them to exchange SQL data with copies in memcached.

Facebook广泛使用MySQL,并将补丁提供给MySQL项目。他们有一个公共Launchpad网站,名为 mysqlatfacebook ,还有一个名为 MySQLatFacebook

Facebook uses MySQL extensively, and contributes patches back to the MySQL project. They have a public Launchpad site called mysqlatfacebook and also a Facebook site called MySQLatFacebook.

对于分布式数据,他们使用一种称为 Map / Reduce ,包含在 Hadoop 项目。他们构建了一个名为 Hive 的项目,以类似SQL的语言查询数据。他们还为特定系统使用其他专门的数据管理技术。

For distributed data, they use a non-relational technology called Map/Reduce, included in the Hadoop project. They built a project called Hive to query the data with an SQL-like language. They also use other specialized data management technology for specific systems.

由于他们总是在创新,必须扩大和扩大规模,他们会定期发明或采用其他技术的数据

Because they are always innovating and having to scale up larger and larger, they regularly invent or adopt other technology for data management.

查看:

  • Data Management, Facebook-style
  • Hive - A Petabyte Scale Data Warehouse using Hadoop
  • Presto: Interacting with petabytes of data at Facebook

这篇关于Facebook墙的数据库结构的文章就介绍到这了,希望我们推荐的答案对大家有所帮助,也希望大家多多支持IT屋!

查看全文
登录 关闭
扫码关注1秒登录
发送“验证码”获取 | 15天全站免登陆