> For the complete documentation index, see [llms.txt](https://eric-zhang-seattle.gitbook.io/mess-around/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://eric-zhang-seattle.gitbook.io/mess-around/overview/estimation_basics.md).

# Estimation

### Performance

* Query with index should be around 1ms \~ 2ms
* One write should be around 5ms for SSD disk

### CPU capability

* 1 CPU core can handle 200 operation
* Usually database server: 56 CPU cores -> 60 CPU cores or more
* 5-10 CPU cores should be enough without cache
* One database should be good enough to handle the load

### Cache performance

* Memcache or Rediss usually are cluster, and usually one operation takes 0.1ms or less.
  * 1 CPU cores => 5000-10000 requests
  * 20-40 CPU cores => 200K requests (one machine)

### QPS and concurrency

* Defined inside Little's law
* For example, if a server has a QPS of 20K and 10ms average response time.
* Then the concurrency number will be 200.