twitter 的一些运营经验

http://assets.en.oreilly.com/1/event/29/Fixing%20Twitter_%20Improving%20the%20Performance%20and%20Scalability%20of%20the%20World's%20Most%20Popular%20Micro-blogging%20Site%20Presentation.pdf

 

Key points:

 

• Databases not always the best store.
• Instrument everything.
• Use metrics to make decisions, not guesses.
• Don’t make services dependent
• Process asynchronously when possible

 

很多地方非常值得借鉴

 

里面提及到的两个东西：

- Kestrel (Memcache-like message queue)  用scala开发的一个MQ服务
http://github.com/robey/kestrel
- mod_memcache_block (Apache 2.x Limiter/blocker) 一个apache module，根据规则来block请求的
http://github.com/netik/mod_memcache_block